The Peter Attia Drive - #257 ‒ Cognitive decline, neurodegeneration, and head injuries： mitigation and prevention strategies, supplements, and more ｜ Tommy Wood, M.D., Ph.D.

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

00:00:15.500 my website, and my weekly newsletter all focus on the goal of translating the science of longevity

00:00:19.820 into something accessible for everyone. Our goal is to provide the best content in health and

00:00:24.780 wellness, full stop. And we've assembled a great team of analysts to make this happen.

00:00:28.900 If you enjoy this podcast, we've created a membership program that brings you far more

00:00:33.320 in-depth content. If you want to take your knowledge of this space to the next level,

00:00:36.920 at the end of this episode, I'll explain what those benefits are. Or if you want to learn more now,

00:00:41.760 head over to peteratiyahmd.com forward slash subscribe. Now, without further delay,

00:00:47.780 here's today's episode. My guest this week is Dr. Tommy Wood. Tommy is an assistant professor

00:00:54.420 of pediatrics and neuroscience at the University of Washington. Tommy's research interests include

00:00:59.740 determining how multiple types of brain injuries can impact brain health across lifespan,

00:01:04.420 as well as developing easily accessible methods with which to track health performance and longevity

00:01:10.480 in both professional athletes and the general population. Additionally, Tommy has acted as a

00:01:15.980 performance consultant for professional athletes in a dozen different sports and most recently worked

00:01:21.380 with a number of Formula One drivers through his work with Hinset Performance, which is how Tommy

00:01:26.120 and I met about five years ago. He also serves as an associate editor of the Wiley Journal Lifestyle

00:01:32.940 Medicine, is a founding trustee and director of the British Society for Lifestyle Medicine, and works with

00:01:38.520 a number of digital health companies for charities that focus on how lifestyle and the environment

00:01:42.980 can affect long-term health and chronic disease. In this episode, Tommy and I focus our conversation

00:01:47.580 on the brain, although we do also pepper in a few conversations around F1, given Tommy's mutual

00:01:54.500 interest with mine. First, we speak about age and age-related cognitive decline. We talk about what

00:02:00.220 cognition and cognitive decline is, including a discussion in depth around memory, reaction time,

00:02:06.360 executive function, memory retrieval, and more. We speak about the root causes of age-related decline

00:02:11.920 and look at cognitive demand, including how we should think about distractions and multitasking.

00:02:17.780 From there, we look at what skills are needed to avoid such a sharp age-related decline and the

00:02:23.020 benefits of different types of brain stimulation. We then talk about different theories on the different

00:02:28.160 types of pathology in dementia and neurodegeneration. Included in this, we speak about the lifestyle

00:02:33.600 factors that can help prevent dementia and the importance of muscle. We also look at various

00:02:38.880 supplements that can help with the prevention and survivability of dementia. We end this discussion

00:02:43.720 talking about head injuries. We speak about what a concussion and a traumatic brain injury are and

00:02:48.680 what the various symptoms are, and ultimately, what are the things that someone can do to minimize

00:02:52.600 the severity of these. So, without further delay, please enjoy my conversation with Dr. Tommy Wood.

00:03:02.900 Hey, Tommy. Good to see you again. It's been, gosh, about six months since I last saw you at

00:03:08.360 Coda, which, of course, we'll talk something about why you and I would run into each other at Coda,

00:03:12.660 given our mutual interest in Formula One. But there might be some people listening to this who

00:03:16.900 aren't familiar with you or your work, though it's certainly been referenced and you've been on a

00:03:20.500 number of podcasts. So, why don't you give us a little bit of your background and we'll kind of go

00:03:23.880 from there. Sure. I do have a varied experience and background, which I think is useful for the

00:03:30.880 bunch of things that I do, but also sometimes slightly confusing because people will know me from

00:03:35.140 one arena, but not know the work that I do elsewhere. I'm an assistant professor of

00:03:39.780 paediatrics and neuroscience at the University of Washington in Seattle. The majority of my work

00:03:46.020 there is in basic animal preclinical research in brain injury. We look at ways to treat the injured

00:03:53.020 newborn and paediatric brain, and we also do some work in traumatic brain injury. But before I got to

00:03:58.540 that point, I trained as a medical doctor in the UK. I worked as a doctor in central London for a

00:04:04.980 couple of years before I did my PhD in physiology and neuroscience. So, though I'm not a registered

00:04:10.620 medical doctor currently, I don't have an active medical license. I do have medical training and

00:04:14.300 that sort of helps inform a lot of the work that I do. Alongside that sort of formal training pathway,

00:04:19.720 I spent a lot of time working with athletes. I was an athlete myself. As a student, I spent some time

00:04:25.080 coaching athletes, particularly rowers. And that was my main sport. And then later on during my PhD,

00:04:31.640 and when I was doing my postdoctoral work, I worked with a company that worked with athletes trying to

00:04:36.980 improve performance or their overall health and their longevity in sport. That was probably the

00:04:41.120 main thing that we really saw a lot of people wanting to focus on. So I have this kind of track

00:04:47.420 alongside where I've worked with athletes in various ways. And then through that, got to working with

00:04:54.080 Formula One drivers in particular through a company called Hintzer. I know you've had

00:04:57.320 our good mutual friend Luke Bennett on the podcast before. So that's kind of where I do some additional

00:05:02.540 work is in athletic performance and health. And in addition to that, also have some interest in

00:05:09.300 long term cognitive functions. So we look at how the brain responds to injury, we look at how to

00:05:15.620 repair that or mitigate injury processes. But what I'm really interested in is how do all these things

00:05:21.000 tie together. So how do aspects around lifestyle and the environment affect how your brain functions

00:05:25.720 throughout your entire lifespan. And so I work with some dementia charities in the UK related to that.

00:05:30.800 And I'm also a founding director of the British Society of Lifestyle Medicine. So I'm particularly

00:05:35.220 interested in how we can use lifestyle to improve population health. Thanks, Tommy. That makes a ton

00:05:41.620 of sense, hopefully for people now to understand the varied nature of both your skill set and your

00:05:46.180 interests. So let's just start by diving into cognitive decline. And I defer to you, Tommy,

00:05:51.400 how you want to do this. Would you like to do this starting with the pathologic cognitive decline

00:05:54.700 vis-a-vis dementia? Or do you prefer to talk about it through the sort of more ubiquitous age-related

00:05:59.360 cognitive decline? So we can start with age-related cognitive decline because that's pretty well

00:06:07.800 described. So if you look across large population sets, you'll see that with increasing age,

00:06:16.180 you see a pretty linear decrease in standardized cognitive function. And that's across all the

00:06:21.700 different types of ways that you can measure cognitive function, aspects of executive function,

00:06:25.560 working memory, except for one type of cognitive function, which is historical memory. And that's

00:06:30.700 probably because of the way that those memories are encoded. They're sort of moved from the main

00:06:35.080 memory storing machinery to, and they kind of spread throughout the cortex. And they're sort of

00:06:40.100 protected from some of the changes that happen as we get older. But in general, you just see this

00:06:44.720 steady decrease in cognitive function as people get older. If you then translate that to what we

00:06:51.660 might call some kind of pathological cognitive decline, which would then lead into frank dementia,

00:06:56.940 which is a long-term loss of significant cognitive capacity or cognitive dysfunction, then you might see

00:07:05.840 an accelerated trajectory. So there's some period of what we call mild cognitive impairment, which you can

00:07:11.240 diagnose with some standardized cognitive tests. And then eventually, that will continue into frank

00:07:17.620 dementia, which there are many subtypes. But the one that probably people are most familiar with and are

00:07:23.320 most concerned about for themselves is Alzheimer's disease or Alzheimer's dementia. And there are probably

00:07:29.200 multiple things that drive both of those paths. But in some individuals, there's this accelerated

00:07:34.800 decline that then ends up in then having the diagnosis of dementia.

00:07:38.440 Tommy, let's go back to the beginning of that and just make sure that we've given people a real sense

00:07:43.340 of what cognition actually is. One of the most common things I hear from my patients is some sort of

00:07:52.040 complaint around memory. Just yesterday, I was talking to a patient and he noted the fact that he had been

00:07:58.760 recently remarried. He was very happy about that, but said, you know, one of the unintended consequences

00:08:04.440 of getting remarried is he just inherited like a hundred new people in his life because all of the

00:08:10.240 folks who, you know, were his wife's side of the family and her friends and stuff are now kind of a

00:08:14.540 part of his life. And he said, I can't remember their names. It's the ability with which I can meet a

00:08:20.580 person, remember their name is decidedly different from when I was 20 years younger. So he's in his late

00:08:26.160 fifties, contrasting this with being in his late thirties. So obviously that's one component of

00:08:31.340 cognition. It, by the way, is hands down the one I hear people most complain about. I don't hear many

00:08:37.120 people complain about decreased executive function, decreased processing speed. I would suspect it's

00:08:44.180 because maybe most people aren't pushing those to their limits and or we don't have as readily available

00:08:50.560 tools to internally discern decreases in that. But can you just expand more broadly on overall this, both the

00:08:58.480 depths of cognition and what it entails, but also this phenomenon that I'm sure the moment someone

00:09:03.620 hears what you do for a living, they're probably right up to you at a party giving you the same

00:09:08.380 complaints. Absolutely. And it's very difficult actually, you know, so you can define these domains

00:09:18.020 of cognitive function. You've essentially already defined them, executive function, which is usually

00:09:22.400 around complex decision-making, but you might, for the average person, it might be, you know, that time

00:09:27.980 when you think about saying something, but then you realize it's a bad idea to say it. That's executive

00:09:33.140 function. That's the, your prefrontal cortex is jumping in and saying, that's a really bad idea, even if it

00:09:38.520 sort of flashes through your mind. But then you have various aspects of short-term and long-term memory,

00:09:43.880 processing speed, reaction time is probably important as well. However, when you talk to individuals

00:09:50.820 about cognitive function, they have their own things that they want to be good at, right? So it's very

00:09:56.860 personal from an individual. Yes, we can, we can use a standardized battery of tests and that's what's

00:10:01.060 done clinically, but there's usually some aspect of function that they notice is declining over time

00:10:06.860 or that they want to be better at. And then they can sort of put focused attention into improving that.

00:10:11.600 And I believe that you can improve that pretty much any stage of life. So that's part of it.

00:10:17.060 And you mentioned memory. And of course, this is something that the people will mention the most

00:10:22.920 and will notice in themselves. But there's actually two different parts to memory. And it's different

00:10:27.980 probably in most people, even in the setting of sort of standard age-related cognitive decline,

00:10:33.880 and in those who have some kind of pathological cognitive decline. The first part of memory is

00:10:39.840 encoding that memory in the first place. The information comes in and your brain signals through

00:10:45.360 acetylcholine and other neurotransmitters to actually say, this is something that we want to

00:10:49.640 recognize and store. And that's the process that seems to be particularly lost in those with

00:10:55.340 pathological cognitive decline. That's why things like coenesterase inhibitors were and are still

00:11:01.480 used in Alzheimer's disease, because that helps to bolster some of those encoding processes, but through

00:11:06.480 acetylcholine signaling. And this takes place predominantly in the hippocampus?

00:11:10.340 Yes. That's where a lot of the process starts. But over time, you get consolidation and these

00:11:16.060 memories may get moved around. Particularly, like we talked about historical memory, they get shifted

00:11:21.200 throughout the neocortex, which is basically the rest of the outside of the brain. The other aspect

00:11:26.660 is retrieval. There's information in there and it's getting it out. And retrieval speed is something

00:11:32.340 that seems to slow down with age. But part of it, you know, often we think of this as pathological,

00:11:38.340 part of it may be that over time, you just accumulate more information. And the more

00:11:42.720 information that's in your hard drive, the harder it is, or the longer it takes to bring out a certain

00:11:47.220 piece of information. Yeah, this is sort of the argument that Arthur Brooks used, that as we're

00:11:51.060 aging, as you add memories, you're creating volumes in a library. And the more volumes in the library,

00:11:57.440 the longer it takes the librarian to go and get the specific reference they're looking for.

00:12:01.420 How can we figure out the relative contribution of library size versus librarian speed when it comes

00:12:10.060 to accessing these memories? Because I guess this is another maybe way to think about that. But

00:12:15.620 the example that resonates for me personally is I either meet somebody and can't remember their name,

00:12:24.300 but five minutes later I can. Or I have an idea, I want to say something about it,

00:12:30.540 and at the last minute I can't remember. But then 10 minutes later, I kind of remember. So

00:12:35.960 it's not that it's not there, but boy, it took me a long time to get it.

00:12:40.340 So I think in reality, it's probably very difficult to pass all of these out. And so I don't think we

00:12:46.920 could pretend that we know exactly the relative contributions. However, some of this is certainly

00:12:52.560 affected by other factors. And that's something that you can take into discussion with, say,

00:12:57.520 individual patients or individuals are concerned about their memory. So it seems like sleep

00:13:02.980 impairment or some kind of sleep deprivation or suboptimal sleep impairs retrieval. So then that

00:13:11.760 could maybe open up a discussion about sleep. Subjective stress seems to also play a role here.

00:13:17.440 So I think some of it is accepting that your library is larger. And some of it is thinking about

00:13:23.840 other factors that may be impairing or allowing for that process to be suboptimal, such that retrieval

00:13:30.140 is harder. Another part that comes into play here, which is also important, and it falls into that same

00:13:35.880 line of thinking, is that as your library gets bigger, your librarian becomes more selective in

00:13:43.540 terms of the things that they want to actually put on a bookshelf. So imagine as you've met hundreds of

00:13:51.940 people in your life, thousands of people, you add 100 new people, it's very easy to say,

00:13:58.120 do you know what, the first time I meet this person, I may never see them again. So maybe it's not

00:14:03.300 actually worth encoding that memory, and you become more selective in what actually gets stored. So that

00:14:08.660 may be part of it as well. And these are not necessarily pathological processes. These may be

00:14:12.940 your brain doing its normal job of, well, how do I figure out what's worth storing? And then how do I

00:14:17.260 retrieve what I've decided to store? This may be a question that goes beyond your level of expertise.

00:14:22.400 So I apologize if I'm asking you something outside of the scope. But I guess, what is a memory

00:14:27.520 physically? And why is there a finite amount of storage? So if I have an understanding of why a

00:14:33.080 hard drive is finite, and if I only have two terabytes on a hard drive, and I keep adding video to

00:14:39.960 that, eventually at some point, there is no more storage capacity. I don't think I have enough of an

00:14:45.380 understanding of what a memory is, and why it would therefore have a physical constraint.

00:14:51.840 So there have been, and I will absolutely agree with you, this is beyond my area of specific

00:14:56.640 expertise. I know that this is a topic that is hotly debated, where some people have said that

00:15:02.320 comparing human memory to a hard drive is essentially, it's a complete fallacy. It's nothing

00:15:08.260 like that. We use it because it's something that we can understand. It helps us sort of apply a very

00:15:14.360 complex process to our own thinking and understanding of how our brains might work. But in reality,

00:15:19.420 that is not how memories work. And there shouldn't be a limit on capacity in the same way that there

00:15:24.760 is with a physical hard drive. However, you might still understand that there are probably still a

00:15:33.020 finite number of things that your brain will choose to encode and store for the same reasons that you

00:15:39.620 only want to have the information that's probably maximally useful for your survival, for want of a

00:15:43.740 better way to think about it. And then that puts some constraints on how the system sets up what

00:15:48.540 it decides to store and then retrieve. My eight-year-old son last night was asking me these

00:15:54.760 questions. It's amazing when kids ask questions you can't understand, you can't come up with an

00:15:59.200 answer to. And he was asking me where the memories were in his brain and how they get there. And I'm

00:16:06.460 like, God, these are really good. Like when I was eight, I wasn't thinking of great questions like

00:16:11.200 this. So anyway, it's disappointing that I can't answer my child's questions. Okay. So we've

00:16:17.760 established that as time goes on, presumably two things are working against an aging individual.

00:16:25.360 One of them, not pathological, one pathological. So the non-pathological is you just have a greater

00:16:32.640 reservoir of memories and your brain might be selectively choosing how to prioritize new encounters

00:16:40.320 and new memories with some understanding that the denominator keeps growing and I have to be

00:16:45.900 selective. But there's also, as you said, or I think as you're implying, there probably are some

00:16:50.700 pathological changes. And whether we use the term pathology or not is probably controversial, but there

00:16:56.980 are some age-related changes that are occurring that are also, for lack of a better thinking in our

00:17:05.300 analogy, slowing down our librarian, reducing our librarian's vision, some way that makes it actually

00:17:12.080 more complicated for us to do these things. What do we think is at the root of that age-related

00:17:18.140 decline that is specific to be it retrieval, computational cycles, processes being executive

00:17:26.080 function, all of these things that we would all prize as important pieces of cognition?

00:17:30.180 So the way that I think about it is that we know with aging, we tend to see a decrease in size or

00:17:41.540 atrophy of the frontal and then the temporal, particularly the medial temporal parts of the

00:17:47.780 brain. And the medial temporal lobe is where your hippocampus sits, as well as some parts of the cortex

00:17:53.760 around it that support it, like the parahippocampal gyrus and the entorhinal region. And there are

00:18:01.200 multiple schools of thought of why those areas of the brain may be particularly vulnerable. Some maybe

00:18:08.000 because of their specific function in memory or because they're deeply involved in the initiation

00:18:13.840 and sort of the continuation and structure of sleep, which is obviously very important for memory

00:18:18.560 consolidation and also various processes of recovery and repair. But there's also, if you think about

00:18:25.760 the whole number of things your brain is exposed to, those areas of the brain seem to be particularly

00:18:30.320 susceptible to negative outside influences and then also susceptible to beneficial supportive processes

00:18:37.280 like actually putting greater demand on those areas of the brain such that they respond and increase

00:18:41.920 in their function. So when I think about the various buckets of things that are required for

00:18:49.120 a healthy brain, for want of a better phrase, they are around supply, vascular supply, supply of metabolic

00:18:56.080 energetic substrate. There are important things around structure and function. So this could be

00:19:01.920 structure related to neuronal membranes, so the importance of say DHA, omega-3 fatty acids, which are

00:19:08.480 concentrated in synapses. They're very important for communication between neurons. And then mitochondrial

00:19:14.480 function as an important part of that. And then you might think of actually placing a demand on those

00:19:20.000 structures. So in most aspects of biology, the function of an organ is proportional to the demands

00:19:27.680 placed on it, so you increase capacity. But then that also requires some period of recovery, and that's where

00:19:34.480 sleep and other things come into play. Plus you might want to avoid negative outside factors. So if we

00:19:40.080 think about dementia, we know that there's some risk associated with things like smoking, potentially

00:19:45.920 air pollution, chronic inflammatory, or infectious conditions like periodontal disease seem to be

00:19:50.800 associated with it. So you want to have this supply of substrate, you want to have good function, you want

00:19:58.160 to make sure you allow that area to rest and recover, you want to avoid things that then may impact those

00:20:04.000 processes. And then sort of what I think is driving a lot of this is the amount that we actually ask

00:20:10.000 those regions of the brain to do, which does decline naturally over time based on how we currently

00:20:15.120 structure our lifespan. So of all the things you said there, I think the one that we're going to click

00:20:20.720 on first, I guess, would be this idea of demand and what we ask of the system. So in certain areas,

00:20:28.320 as you point out, it's pretty intuitive. You cannot maintain muscle mass without putting the muscles

00:20:34.640 under significant demand that is so strenuous that you wouldn't be able to maintain it indefinitely,

00:20:41.200 right? If I look at the workout I did this morning, I wouldn't be able to do that to my muscles or to my

00:20:47.040 heart indefinitely. You can do it for a few hours. But as you point out, if nothing else through sleep,

00:20:53.360 but even more than that, there are just days when you wouldn't push that hard. In other words,

00:20:57.280 you sort of think of exercise as a hormetic stress. Now, I haven't thought about it this way,

00:21:01.840 but there are certain organs for which I would guess that that's not true, right? I mean, does

00:21:05.760 the liver need to be stressed? Do the hepatocytes need to feel the insult of ethanol to otherwise

00:21:12.320 perform well? I guess I haven't thought about it through the lens of kidneys and the liver and stuff.

00:21:17.280 What do we know about other organs and their need to do what say the heart does or skeletal muscle does?

00:21:23.600 I've thought about the liver in particular, and I think you can say that the case holds,

00:21:29.600 particularly with alcohol exposure, as the example. That's what you said, and that's what I think of as

00:21:33.680 well. Not that the liver doesn't function without alcohol exposure, but if it wants to

00:21:39.200 optimally deal with a certain type of product, say ethanol, it wants to metabolize it, then we know that

00:21:46.560 with chronic alcohol exposure, before we get to the point where we damage the liver,

00:21:51.440 you see an upregulation in size chrome P450 2E1, you see an upregulation in aldehyde dehydrogenase,

00:21:57.120 you see an upregulation of mitochondrial function and metabolism to regenerate NAD, which is sort of

00:22:03.680 the rate limiting step for alcohol detoxification. So yes, if you stress the liver with alcohol,

00:22:09.120 it will upregulate its function in order to have a greater capacity when the next drinking

00:22:13.600 session occurs. So I think there are sort of parallels across multiple organ systems.

00:22:18.960 Yeah. And just make sure listeners aren't hearing this and thinking, oh, he's telling us to drink

00:22:23.120 more, to drink more. No, I think what you're saying is, and we would all agree that the health benefits

00:22:27.280 of alcohol are none, but you're saying before you get to destroying your liver with alcohol

00:22:33.120 irreversibly vis-a-vis cirrhosis, if your goal is to be able to drink two drinks a day,

00:22:38.560 you have to drink daily. You're going to have a better job tolerating two drinks if you occasionally

00:22:46.000 have a drink, as opposed to if you never have a drink. I mean, that's sort of what you're basically

00:22:50.800 saying. Yeah, that's right. And so if you want an organ system to function in a specific way,

00:22:57.040 so you want your brain to function, improve its function in a specific domain, or you want your

00:23:02.080 body or your skeletal muscle or your cardiovascular system to function better in a specific domain,

00:23:06.240 you want to train for a marathon, or you want to be a capacitive powerlifter, you apply a relevant

00:23:11.840 stressor that's, like you said, it's hormetic, and you give time to recover and adapt to it,

00:23:16.880 and then you get an increased capacity later. And so I think that that's very relevant for the brain,

00:23:21.260 but I might use exercise as a way for people to better understand it, because you can kind of see

00:23:26.440 that happening. But then to kind of draw that analogy out, you might say, okay, there are other

00:23:31.600 organ systems where there's evidence that that's the case as well.

00:23:34.140 So let's go back to the brain and talk about what might be a difference between kind of a positive

00:23:40.720 versus a negative demand. So I'm sure most people listening to this podcast right now are under some

00:23:46.200 cognitive demand. We're not just sitting here idly shooting the breeze. We're talking about stuff that

00:23:51.640 for most of us requires some thought, some concentration to pay attention to this. So

00:23:55.920 is listening to this podcast for different people producing different levels of cognitive demand,

00:24:01.520 for example, depending on their level of familiarity with this subject?

00:24:05.260 Yes, absolutely. And again, I think that the idea of cognitive demand is relative to the individual

00:24:13.060 as well as what they want their brain to function best at. However, when we think about cognitive

00:24:19.880 demand, I think there's multiple different ways that you can come at it. So when I think about

00:24:26.380 generating skills, or maybe just brain development more broadly to start with, you might think about

00:24:34.220 how does a toddler interact with their environment such that they're developing motor skills, language

00:24:41.220 skills, social skills. And it's often this concerted effort for a short period of time, where you are

00:24:49.660 right at the limit of your current capacity, being able to stand, being able to walk, being able to climb a tree,

00:24:54.800 being able to pronounce a certain word. And then maybe you could put a timeframe on it, maybe it's

00:25:00.520 somewhere between 20 or 30 minutes, something like that. Adults may be able to do this for longer.

00:25:05.280 Right at the edge of your skill set. And then probably there's some failure in there, because

00:25:09.740 the process of failure sort of upregulates the processes of focus and attention as well, up to a

00:25:14.620 point before we get frustrated. And then there's some rest and recovery, your toddler is going to sleep a

00:25:20.700 bunch after they spent a lot of time exploring the environment, trying to improve their motor skills.

00:25:25.160 And I think that kind of provides an idea of the type of focused attention that you put into

00:25:34.140 something that then drives plastic reorganization, which is what we care about. We want to try and

00:25:39.260 drive an increase in functional capacity in some domain of cognitive function. And so it's probably

00:25:45.240 going to be something that looks like that. And there are benefits to continually doing things at the

00:25:53.460 level of your current capacity, right? If you're an athlete, you're not always pushing the boundaries, you're

00:25:58.960 not always doing a red line session that drives a bunch of adaptation. And I think the brain is similar. But that's

00:26:05.340 very different from how most adults perform in their day to day work, where yes, it's very it feels

00:26:12.960 cognitively demanding. But you're multitasking task switching. So you're never providing focused

00:26:19.580 attention on a specific subject. And you're also not really providing that stimulus to increase skill in a

00:26:27.740 specific area necessarily. Even though you feel busy, even though something feels cognitive demanding,

00:26:32.340 I don't think that those stimuli are the same in terms of what's driving a functional change in some

00:26:38.120 area of the brain. Yeah, it's a very important point you raise. And it, of course, always begs a

00:26:42.640 question that I have when people tell me, or when I feel this way myself, that I just can't remember

00:26:49.140 things. I'm just not as facile cognitively. And I always just wonder where distraction and lack of

00:26:56.240 focus fits into the mix. So are there any kind of heuristics for what is too much task switching?

00:27:02.860 What is too much distraction? Because as you said, you feel very busy sometimes. And I actually went

00:27:08.460 through a little bout of this yesterday. I was really trying to get a lot of things done. And then one

00:27:13.200 more thing got put on my plate. And I had this window of 10 minutes where I accomplished literally

00:27:20.500 nothing by toggling back and forth between three different emails, text, WhatsApp, and a document I

00:27:29.400 was trying to work on. And I just couldn't get anything done. I mean, to be that debilitated is

00:27:33.460 pretty unusual for me. But in that 10 minutes, I accomplished exactly zero. I would have been better

00:27:38.060 off for 10 minutes literally just walking around the house or walking around the block.

00:27:42.320 So clearly there was sort of a threshold there. But do you have any way to think about what that

00:27:46.460 looks like and where you've crossed a line into being busy and unproductive rather than busy and

00:27:52.020 productive or focused and productive? I would probably make the argument that humans in general

00:27:58.500 cannot multitask in the way that you describe it. And again, we can sort of break this out. There

00:28:05.620 are probably two different types of multitasking. One is the automatic performance of learned

00:28:11.500 subroutines that kind of just happen. And you could do multiple of them at the same time. So say

00:28:16.540 you're a dancer, and you're running subroutines, the steps you've learned, interacting with a partner,

00:28:22.280 listening to the music, you are technically multitasking. But those are all learned subroutines

00:28:27.120 that are essentially happening automatically or subconsciously. What you're describing as

00:28:32.760 multitasking is that process of focusing on one thing, then another thing, then another thing. And we know

00:28:37.460 that there's a loss function in terms of the time it takes to get back onto a new task from the previous

00:28:43.040 task. And it's something in the order of, you know, like, people have said 20 something seconds,

00:28:47.400 right, for you to refocus onto something. So say you're switching your focus every minute or two,

00:28:54.820 the amount of time you actually have to focus on one specific task is dramatically reduced, because

00:28:59.320 a significant proportion of that is just spent with your brain figuring out what it is you're asking it

00:29:03.560 to do. So in that setting, say you're doing things that allow you to enact, learn subroutines in your

00:29:12.260 work, you could probably do that continuously. And there's less of this sort of stressful demand

00:29:18.160 on your attention. But if you're needing to directly focus on multiple things at the same time, then

00:29:23.980 switching from one to the other, I would argue that the majority of people cannot do that well,

00:29:28.880 even though they may think that they can.

00:29:31.120 Yeah. One way that I think about this in terms of exercise is if I'm doing a steady state aerobic

00:29:37.460 efficiency, we call it a zone two workout on a stationary bike, I can listen to a podcast and an

00:29:43.700 audio book and be completely focused on it. I can still manage to be on a bike and pedal.

00:29:48.480 I don't have to worry about traffic or anything like that. And I can be focused. But if I'm doing a

00:29:53.440 higher intensity cardio workout, or if I'm lifting weights, I can't listen to podcasts and books.

00:29:58.900 Those tasks just demand a little bit more of my attention, either to not get hurt or just to

00:30:05.360 focus on the movement. In other words, they're not as presumably automatic as just holding 90 RPM at a

00:30:12.220 fixed wattage riding around. Would you say that that's kind of an example of something that's

00:30:17.380 really automatic versus something that's just a high enough order of processing where you can really

00:30:20.920 only do that one thing? Yeah, I think that's a good way to think about it. And more broadly,

00:30:27.100 I think that traditional work based multitasking is probably the point where there's this biggest gap

00:30:33.980 between perceived demand and the amount of beneficial cognitive stimulus you're actually getting.

00:30:42.380 And it reminds me of something that a former colleague of mine at Hintzer, James Hewitt,

00:30:46.720 he also used to be a professional cyclist, and he calls this the cognitive middle gear.

00:30:52.380 And it's this point where effort is high, but sort of the end product is minimally useful.

00:31:00.640 If you can think about athletes who may go out and thrash themselves at threshold for like an hour

00:31:08.020 and do that every day, it's very hard. It feels really hard. But in terms of physiological adaptations

00:31:15.700 that improve performance, there's a big gap between how hard it is and benefit. And I think

00:31:20.680 this kind of multitasking is the same. Yeah, it's a physiologic no man's land.

00:31:25.060 Exactly. Yeah. Yeah. It's too hard to give them the wide aerobic base and not hard enough 0.97

00:31:30.320 to give them that peak performance. Well, this is actually kind of a nice quick way to segue,

00:31:35.280 and I want to come back to this, but just let's talk about Formula One for a second. So

00:31:38.400 you and I have been fans of this sport for a very long time, but I think a lot of people listening to

00:31:43.280 this have become fans of F1 through Netflix's Drive to Survive series. Of course, if you're

00:31:48.660 listening to this and you don't know what Drive to Survive is, I recommend you go back and watch the

00:31:52.740 last two seasons at least. But I think anybody who's watched the sport will appreciate how much

00:31:59.380 they need to be able to do while driving. And on a personal level, this is something I can relate

00:32:04.980 to because I spend a lot of time in a car and in a simulator, and I am nowhere near being able to do

00:32:12.980 what a driver does. So for folks to get a sense of this, first of all, they're traveling at speeds

00:32:18.860 that are simply unbelievable, right? Their straight line speed is 200 to 220 miles an hour,

00:32:25.560 depending on the setup they have for that race. And the amount of things that they have control over

00:32:31.120 on this computer that masks arrayed as a steering wheel in front of them is unbelievable, right? So

00:32:37.060 between every corner, they're making some adjustment. They're switching the brake bias with one knob,

00:32:42.980 which shifts the emphasis or force between how much brake is on the front wheels versus the rear.

00:32:49.160 They're altering the slip angle of the differential, which is allowing inner wheels versus outer wheels

00:32:55.200 when they're going around corners to move at different relative speeds. They're obviously

00:32:59.440 adjusting drag reduction system when appropriate. They're adjusting access to battery power if they're

00:33:04.900 passing or overtaking or trying to defend. There's simply no shortage of things. Oh, and by the way,

00:33:10.280 they're pushing a radio button to talk to their engineer who's talking to them the whole while.

00:33:14.760 So I will just say, Tommy, I have been driving for nine years. I can't come close to managing all of

00:33:22.040 these variables in the simulator even, let alone, I mean, the most I can handle when I'm driving

00:33:27.540 is driving and talking and maybe adjusting brake bias here and there, but I'm just so beyond my

00:33:36.880 capacity to do any more. So how do these guys do so much? And I think more importantly, for those of us

00:33:44.060 who can't, what do we need to do to get better at that capacity? Or is it just so sports specific?

00:33:52.180 And because these guys have been driving carts since they were five, things that I have to think

00:33:57.260 about while driving, they don't. Is that simply what it comes down to?

00:34:00.040 I honestly think that's a lot of it. Many of the things they're doing, like we talked about this

00:34:05.800 idea of multitasking being running several learned subroutines at the same time, which the human

00:34:11.020 brain is very good at. That's essentially what they're doing for a lot of those functions. And

00:34:16.900 if you've spent your entire life as a racing driver, which they have essentially, they've been

00:34:21.920 carting since they were young kids, these guys, then when the computer setup changes, or, you know,

00:34:28.060 there are new things on the car based on new regulations, right? They're just stripping off

00:34:32.540 a top layer and adding on a new one. So what they have to learn is very minimal compared to

00:34:37.560 everything else that's probably happening automatically. So yes, I mean, there's a huge

00:34:40.640 cognitive demand. And then probably a lot of what they're actually using their brain power for in

00:34:46.880 the moment is reacting to others, right? You can't learn that.

00:34:50.000 Yeah, it's racecraft.

00:34:51.060 Yeah, exactly. And so that's where most of their attention is, and everything else is happening

00:34:55.360 automatically. Now, when you have drivers that move up the ranks, and there's a pretty big jump

00:35:01.440 from F3 to F2, from a complexity of the car standpoint, and again, from F2 to F1,

00:35:08.320 are there things that coaches will do with the drivers during those big transition years

00:35:12.880 to really help them get up to speed? I'll give you an example. I don't think this is an example of

00:35:18.060 what I'm accomplishing, but just people are probably very familiar with watching the drivers

00:35:22.000 before races play sort of half physical, half cognitive games, right? Where they have like

00:35:28.340 little lights on a board and they have to touch them or they're playing catch or juggling or doing

00:35:32.780 all these sorts of things. Are those things doing anything to sharpen their brain or do they just

00:35:37.420 look cool? Probably a bit of both, but I would argue that for some drivers, it certainly seems to be

00:35:43.400 beneficial and it's very different from driver to driver. However, there are definitely drivers who,

00:35:49.020 when they move up the ranks or even those who've been in Formula 1 for a long time, they'll have

00:35:53.400 a dummy steering wheel, right? So they can move through the patterns of doing everything,

00:35:59.500 even if they're not on track. And this is often the same as when we're learning subroutines. So

00:36:04.280 say you're learning a new dance, you'll do the steps one by one without music, then you'll start to

00:36:10.080 string them together, then you'll add complexity, and it's the same thing. So there are multiple

00:36:15.440 opportunities for them to practice some of these skills, even though they don't get a huge amount

00:36:19.360 of time actually racing on the track. When you're thinking about the sort of reaction games that some

00:36:26.020 drivers are doing, and some of them may do it in terms of, you know, practicing starts. And again,

00:36:30.820 they'll have sort of a dummy setup to do that or balls and reacting to somebody throwing something.

00:36:36.260 Some of it may be placebo, right? It helps them feel like they're getting ready, which is great. And

00:36:40.720 anything you can do to improve that, you know, you would welcome. But then the other part of it may

00:36:45.440 be moving you onto the appropriate part of the Yerkes-Dodson arousal curve, right? Which basically

00:36:51.040 says that there's this U-shape or inverted U-shape curve of like how aroused you are and how you perform.

00:36:57.000 And that curve is different based on the sport. And so some period where you're having to focus and

00:37:02.840 react, but also remain relatively relaxed, because that's incredibly important, right? You want to be fast

00:37:08.620 off the start line. But then the first thing you have to do is get into turn one when you have

00:37:11.820 eight guys around you trying to navigate at the same thing and you have to react to them. So if

00:37:16.440 you were very tense, that would decrease your performance in that setting. So I think some of

00:37:21.360 it is just getting you to the point where you're focused, but then also relaxed, you know, if you're

00:37:26.360 doing it with some kind of skill-based thing that requires you to also be relatively fluid. So it can be

00:37:31.940 like catching a ball suddenly, right? You can't do that if you're very stiff and tense. So I think some of it's

00:37:36.980 this balance of getting into the right mindset before you then have to do something once you get

00:37:41.280 into the car. Now, most of us, we're never going to be professional drivers or athletes. And certainly

00:37:46.520 as we're aging, the need for really complex motor skills and cognitive skills may go down. But what

00:37:54.620 is the equivalent that you think about for a person as they reach middle life and they're thinking about

00:38:00.920 transitioning, you know, or what lies ahead of them as they transition into older age? In other words,

00:38:05.500 what is that apex set of skills that they need to be able to have to give them the highest amount of

00:38:13.680 physiologic headroom possible to avoid or minimize this age-related decline? And more importantly,

00:38:21.300 potentially even avoid the pathologic stuff that we haven't talked about yet, but we'll come to.

00:38:25.800 I think a lot of it, again, is probably quite individual. If you want to be able to perform a

00:38:34.060 specific task, you want to be able to drive cars for as long as you're able, right? The next four or

00:38:40.520 five decades. So then you would want to push your skills as far as you can in that arena while you're

00:38:48.300 able. And then you have, like you said, maybe we'll talk about more about this idea of headroom.

00:38:53.240 You cannot, at least not yet, we cannot completely stop the aging process. But you may be able to slow

00:38:59.880 the decline and or if you increase your level of capacity, it will take longer to get to a point

00:39:06.060 where function is lost such that you can no longer engage in that activity. When you're thinking about

00:39:11.300 the brain, there's a whole bunch of things related to language skills. Obviously, memory is important,

00:39:19.260 but how you interact with your environment. And I think social interaction is critically important

00:39:24.100 and something that is probably under-discussed in relation to long-term cognitive function. That's

00:39:29.500 a critical aspect as well, particularly as people spend less time with others because of societal

00:39:34.720 effects. But when you think broadly about how cognitive function declines with age,

00:39:43.280 it seems to mirror very closely the amount of demand that we put on our brains. And again,

00:39:51.120 how society is constructed. Because cognitive function essentially increases from birth to some

00:39:59.560 peak in late teens or in your early 20s, which is the period of formal education. It is your job

00:40:07.440 to learn. It's your job to develop skills. That's when most sports are learned. That's when languages are

00:40:13.800 learned. That's when skills are learned. And then after that, you essentially spent a bunch of time

00:40:22.280 doing the same thing over and over. Or you become hyper-specialized in one specific skill, and that's

00:40:27.920 rewarded in a number of jobs, right? And so say you're a surgeon, you want a lot of those processes

00:40:32.620 to be automatic. You don't want to have to think about all of them continuously. So it's beneficial,

00:40:37.040 makes you better at your job. But there's much less room for that period of skill building or putting

00:40:45.520 in effort into developing or providing those kinds of stimuli that then drive plastic reorganization

00:40:51.760 in the brain and may increase headroom. So I think some of that natural decline with aging is a function

00:40:59.900 of how we use our brains in general in society. And then there's a drop-off when we retire, and we can

00:41:06.640 see that in various different types of data sets where those who retire earlier seem to experience

00:41:12.100 cognitive decline sooner. And that's probably because the cognitive demand that we do get

00:41:16.240 in our daily lives, the majority of that comes from work. So an important answer, I think, is,

00:41:22.540 if you're trying to maintain a basic set of cognitive functions, is to actively work on ways to increase

00:41:29.960 headroom, increase absolute capacity throughout the lifespan. Because, I mean, at some point,

00:41:35.780 capacity will decrease. But you want to push that out as far as you can. Hopefully, you'll die of

00:41:41.280 something else before you lose the majority of your cognitive capacities. 0.83

00:41:45.760 Yeah. I mean, again, I think everybody's aware of the anecdote, right? Where, boy, you know,

00:41:50.740 Sally was just sharp as a tack, and then she retired, and all of a sudden, it all went to hell in a

00:41:55.660 handbasket. But you hear this so many times that you realize there must be something to this. It can't

00:42:03.120 just be an observational phenomenon that's best explained by something else. There may be other

00:42:08.620 contributors to it. Maybe people who are retiring younger also have more health challenges. Maybe

00:42:14.400 they're of lower socioeconomic status. I mean, you could come up with a lot of confounders that could

00:42:17.980 explain this. But I suspect that there is also a signal there. There's some fire in the presence of

00:42:23.920 that smoke that says, if you retire and, in its place, add nothing cognitively, you could expect

00:42:32.400 to see a decline. I also can't help but wonder how much of this has to do with sense of purpose,

00:42:37.840 which, again, I think maybe falls outside of, we're getting really warm and fuzzy outside of

00:42:41.880 the scientific discussion. But the question I always have is, look, retirement should be thought

00:42:46.920 of maybe as a financial decision. Maybe retiring means I no longer need to work for money,

00:42:51.800 but I'm going to work on something else. And if that something else is not as cognitively demanding,

00:42:57.720 right? Let's say you go from being an accountant where, you know, it's pretty cognitively demanding.

00:43:01.880 You're in a spreadsheet all day. And you say, well, look, I'm 65. I'm done with that. And I don't

00:43:06.140 need to work anymore. But now I'm going to go and do something philanthropic where I'm going to work

00:43:10.340 for, I'm really interested in homelessness as an example. I'm going to go and do X, Y, and Z.

00:43:14.560 You probably have more sense of purpose. You might derive more satisfaction from that,

00:43:18.360 even though it's not as cognitively challenging. Do we have any sense of how that factors into it?

00:43:23.620 Or is that just so far outside of our ability to kind of understand risk?

00:43:28.420 The majority of studies that have looked at this, I guess they fall into two camps,

00:43:33.240 which partially answer your question, but don't necessarily answer it fully. The first is looking

00:43:39.440 at the removal of that stimulus through retirement. It's been done in several population-based

00:43:44.760 studies. They usually account for medical conditions that might cause you to retire early

00:43:50.800 because that's an important confounder. And when you look at other studies, there's evidence

00:43:57.780 to suggest that late in life cognitive activity. So whether you regularly play chess or you dance

00:44:06.300 or you do something else that's cognitive stimulating, that's protective against incident

00:44:10.720 dementia or cognitive decline. So the two parts of that would say that removal of work

00:44:18.380 as a major cognitive stimulus increases risk, but that adding some other kind of cognitive stimulus

00:44:26.280 mitigates that risk. And there are several studies where you do some kind of cognitive training.

00:44:32.540 Maybe it's a computer-based brain training in older adults in their 70s. You can see significant

00:44:38.900 improvements in cognitive function. And if you think about all the things that are most protective

00:44:43.920 in terms of preventing cognitive decline, there was a big meta-analysis done by Jintai Yu,

00:44:50.340 who's a professor in Shanghai, looking at all the different potentially modifiable factors for

00:44:55.140 cognitive decline. The two most important protective ones were early in life education,

00:44:59.560 which I think of as increasing headroom. So the more you learn and skills you develop early in life,

00:45:05.940 and the longer you do that for, the greater headroom you have. And then late in life cognitive activity,

00:45:11.040 which then provides that protective factor. So I think there's enough evidence, as much as we can

00:45:15.660 right now, and most of this is observational, although there are some interventional trials in older

00:45:19.380 adults. You can say that if you're no longer working, if you replace it with cognitively stimulating

00:45:25.040 activities, you're probably mitigating all of that risk, or at least most of it.

00:45:29.320 And we've discussed this a little bit in the podcast in the past, that not all cognitive

00:45:32.740 tasks are created equal. So for example, my reading of the literature was there wasn't any evidence

00:45:39.360 that doing crossword puzzles is necessarily going to do anything for you other than make you better

00:45:44.620 at doing crossword puzzles. But dancing, now that's a bit different actually, because no two steps are

00:45:51.440 ever exactly the same. You're always sort of problem solving, especially if it's sort of complicated

00:45:54.700 dance. Solving a business problem is more elaborate than doing Sudoku or whatever it's called,

00:46:00.500 where they're just sort of little predictable word games. And so would you agree that maybe

00:46:06.240 what you replace it with, there's also some variability in the complexity of that? And the less

00:46:12.040 color by numbers or paint by numbers it is, the more likely it is to... Well, I mean, let's use that

00:46:17.180 example. Painting by numbers versus painting. You're both painting, but in one case, it's a much higher

00:46:23.380 level of cognitive load. Yeah. And I think there were two streams of evidence that maybe support

00:46:31.600 this. So one is around cognitive activities performed on a computer. So there are a number

00:46:37.760 of different ways to do online brain training. There's a system called Brain HQ, which probably has

00:46:43.440 the best evidence to support. It was actually developed by one of the researchers who sort of

00:46:47.920 did a whole bunch of the primary basic research in terms of how we learn in the first place back in

00:46:53.300 the 90s. They have some nice data that shows that if you do these complex training games, which are

00:46:59.260 often reacting to something or shifting focus, right? They require you to be... They're very

00:47:05.380 interactive. Then you can see parallel improvements in things like verbal memory and executive functions.

00:47:11.560 So things that you actually care about in real life. You're not just getting better at the game,

00:47:14.700 you're getting better at certain cognitive functions, which is what you really care about.

00:47:18.060 And that program is called what? Brain HQ.

00:47:20.860 And that's something that people would do online. That's like a game you would play online.

00:47:25.320 Yeah. It's a subscription service. I have no relationship with it, but in terms of online

00:47:29.700 brain training systems, it's been used in a bunch of clinical studies. It's probably the best evidenced

00:47:35.660 one in terms of creating sort of translational improvements in function. And then related to that,

00:47:41.660 there are also studies using video games, where if you randomize people to play solitaire versus

00:47:48.780 Angry Birds versus Mario 3D, the 3D game results in better improvements in working memory. And there

00:47:56.140 is some response inhibition, which is a version of executive function. So the more complex, the more

00:48:01.420 interactive something is, seems to be the greater the improvements in cognitive function associated with

00:48:05.660 it. Then related to that is work in physical activity. So you mentioned dancing, there are

00:48:12.620 studies that have compared dancing to circuit training that is as cardiovascularly challenging,

00:48:21.180 but obviously without that element of social interaction, music, movement, steps, reacting,

00:48:27.100 and you see better improvements in the dancing group. And some people have termed this open skill versus

00:48:32.700 closed skill, physical activity. So closed skill is unidirectional, doing the same thing again and

00:48:38.700 again. So like sitting on an exercise bike inside, you're not doing anything else at the same time,

00:48:42.780 except maybe listening to a podcast versus say table tennis or badminton. So the physical nature

00:48:49.580 of it, the cardiovascular stimulus is the same, but you're not reacting to the environment and other

00:48:54.220 people. And when you do those open skill type of physical activities, you seem to see some greater

00:48:59.420 improvements in cognitive abilities. So there are those various things that say that the more domains

00:49:05.500 or the more complex, the interaction associated with the activity, the greater the associated

00:49:11.100 improvements. Yeah. The more variability there is, right? It's, you know, to be able to walk outside

00:49:16.220 on an uneven surface where the slope is constantly changing and you kind of have to be able to physically

00:49:23.500 and cognitively be aware of what's beneath your feet. It's going to be a lot better for you than walking

00:49:28.060 on a treadmill or even walking around a track in circles. So let's talk a little bit now about the

00:49:32.620 pathology side of this thing. So again, I think listeners to this podcast have a decent understanding

00:49:37.500 about Alzheimer's disease and Alzheimer's pathology. We've certainly had a number of podcasts on the

00:49:40.940 topic, but maybe let's just kind of refresh people's memories on the difference between Alzheimer's

00:49:47.420 disease specifically and perhaps other kinds of dementia, such as vascular dementia, frontotemporal

00:49:52.620 dementia, Lewy body dementia for that matter, and maybe even forms of dementia that don't

00:49:57.420 necessarily fit neatly into these boxes. Like you said, those are probably the four

00:50:02.220 main types of dementia, although there are other ones that then are associated with maybe other

00:50:07.020 neurodegenerative conditions. So the types of dementia associated with Parkinson's disease or ALS,

00:50:11.340 they can be sort of complex and multi-domain and you might see changes in different areas of the brain.

00:50:16.860 The main thing that ties together Alzheimer's disease, and this is something that's worth digging 0.82

00:50:22.140 into because I think both the genesis of the eponym as well as how we now use it is very interesting.

00:50:29.900 But the thing that kind of ties together forms of Alzheimer's disease, and there are two as we

00:50:34.940 think about them, early onset Alzheimer's disease and late onset Alzheimer's disease.

00:50:39.980 But what really ties them together is the neuropathology, which is if you slice somebody's brain open,

00:50:45.740 particularly again within the medial temporal lobe, that's where the primary

00:50:49.980 atrophy and pathology seems to exist, although it can be throughout the brain, is amyloid plaques

00:50:55.820 and tau tangles, hyperphosphorylated tau, within the neurons. And this was how the disease was

00:51:04.700 originally classified. There was an initial case, August D, who Alzheimer treated in an asylum for several

00:51:13.260 years, and then looked at her brain after she died, and he saw these things under a microscope,

00:51:18.620 then collected other cases where they saw similar things. And this was done right at the beginning of

00:51:25.100 biological psychiatry, where trying to find biology that explained psychiatric symptoms.

00:51:32.220 And one of the things you could do at the beginning of the 20th century was look at things under a

00:51:36.140 microscope after that person had died, and that's how they classified it. However, whenever you want to,

00:51:42.380 we can get into how actually these pathological hallmarks correlate very poorly with somebody's

00:51:50.380 symptom burden and disease progression. And the reasons why they accumulate may be very different

00:51:57.580 from person to person and may matter much less from one person to the next. However, it's these

00:52:02.940 pathological hallmarks that created the classification of what we call Alzheimer's disease.

00:52:07.900 You know, it's funny, I wrote about this story in my chapter on Alzheimer's disease and pointed out

00:52:14.540 that many years later, they exhumed, I guess, part of her brain. And lo and behold, she actually didn't

00:52:21.020 have the sort of typical Alzheimer's disease that we see today, which is the disease that 99% of people

00:52:28.060 with Alzheimer's disease get. A very small subset, she was in that subset, get a variant of the disease

00:52:34.540 that is genetically predetermined. And I can't remember which one she had. I think she had PSCN1,

00:52:40.380 or did she have APP? It was PSCN1, but that's actually quite hotly disputed. There was a paper in

00:52:47.740 Lancet Neurology 2013 where they sequenced her brain and they supposedly found this mutation. But then another

00:52:53.740 group got another sample of her brain and couldn't find any mutations associated with familial or early

00:52:59.340 onset Alzheimer's disease. Though if she did have early onset, a monogenic, also an adult mutation,

00:53:05.420 some people have looked into her family history afterwards and her children didn't seem to get it.

00:53:09.180 So even if she had a spontaneous mutation, it didn't seem to be passed on. So there are a lot

00:53:14.060 of people who think that actually there's no evidence that she had a mutation. That may be the

00:53:19.100 first correction I need to make in the second edition of my book because I noted that she in fact

00:53:24.380 did have it, her PSCN1, and as a result it may be a different disease entirely. It may be that PSCN1

00:53:30.380 mutation, PSCN2, APP mutations, which are all these essentially autosomal dominant deterministic

00:53:36.860 mutations that result in people getting this disease and getting it very early. I mean, this is

00:53:41.660 not uncommon for someone to be stricken in their 40s and certainly be dead before they're 60.

00:53:46.860 What is the alternative explanation? If in fact she did not have any of those mutations,

00:53:51.180 what is the best case explanation for her disease, its severity, and the histologic findings?

00:53:59.500 By some telling of the story, after examining her brain, Alzheimer's was encouraged supposedly by his

00:54:08.380 mentor, Emil Krappelin, to gather together cases to sort of build this idea of this common pathologic

00:54:15.500 process. And again, it's sort of, you know, you're recounting history, so it depends on who's telling

00:54:21.180 it. But some people say that actually Alzheimer was quite reluctant to try and group these people

00:54:25.420 together because they were so different. And I know you've sort of mentioned that famous line,

00:54:30.780 which is that once you've seen one case of Alzheimer's, you've seen one case of Alzheimer's,

00:54:34.380 right? They're also very different. And apparently this is something that Alzheimer felt as well. He wasn't

00:54:38.380 sure that they should be grouped together. In the version of the story where she doesn't have an

00:54:44.700 autosomal dominant mutation in PSCM1, it's that she did have some kind of decrease in her cognitive

00:54:52.380 faculties. There are a whole bunch of environmental other factors that could play into that. We could

00:54:56.460 certainly talk about that. But it led her husband to then put her into an asylum. And we know that

00:55:03.420 one of the fastest ways to trigger cognitive impairment or cognitive decline is to basically

00:55:08.220 remove somebody from their environment and completely remove all stimulus, social interaction,

00:55:13.820 and all the things that ground them in, you know, who they are. So that version of the story says that

00:55:19.020 because of the asylum that she was put in, and the way that she was sort of completely disconnected from

00:55:23.660 her normal environment, that then triggered an acceleration in her cognitive decline, which then,

00:55:29.020 you know, maybe a sort of maximized version of what we might see in individuals nowadays who

00:55:35.260 experience cognitive impairment and then dementia. Do we know her APOE4 status when they were looking

00:55:40.620 for PSCM1? Did they check APOE? That's a good question, but I don't know if they checked that,

00:55:46.460 or if anybody's checked that, or if they have, I haven't seen it. Let's sort of talk about the

00:55:50.140 current state of affairs with Alzheimer's disease. And let's go back to a point you made earlier.

00:55:54.060 Do you know what the discordance is between the presence of amyloid beta on a histologic sample,

00:56:04.460 obviously taken post-mortem, and the presence of and or even severity of dementia-related symptoms

00:56:13.740 in that person while they were alive, and the contrapositive of that, which is to say

00:56:18.700 the severity of symptoms in a person while they were alive, and the absence of amyloid beta on a

00:56:25.900 pathologic specimen after they've died? Because in an ideal world, it would be a one-to-one mapping

00:56:30.380 that is 100% concordant. Anyone who has symptoms will have amyloid beta on autopsy, and nobody who

00:56:38.780 doesn't have symptoms will, and anybody who doesn't have symptoms never has it, and nobody who doesn't

00:56:44.460 have symptoms will have amyloid, right? It's a perfect one-to-one mapping. Well, it's clearly not

00:56:48.300 that. So how messy is it? It would be nice if we could put an r squared or an r value on this to see

00:56:54.700 how tightly correlated they are. I think, you know, I have to reach deep into the depths of my memory,

00:57:00.540 and if people have tried to do that kind of correlation, it's, you know, the r is somewhere

00:57:04.540 around 0.1 or something. Which just means for people it's virtually uncorrelated.

00:57:09.900 Yeah, there's a couple of percentage points, maybe, in the variability in cognitive function

00:57:14.060 that's explained by the variability in all the amount of amyloid that you have. And that's very

00:57:18.780 clearly described in animal models as well as in humans. And we also, there are several studies

00:57:24.700 where they've made some very good drugs that can decrease plaque burden in the brain, but that

00:57:28.700 doesn't seem to then correlate with later cognitive functions and may have a high risk of side effects,

00:57:34.540 although that's a whole other story that we won't, don't need to get into today. So

00:57:38.620 pretty much anybody who dies with dementia or experiences dementia will have some burden of

00:57:47.980 amyloid plaque and tau tangles. I don't think I know of a case where somebody had that without any,

00:57:54.460 but we also know that these things naturally accumulate over time. And you can have people,

00:57:59.660 you know, there are multiple cases where they've looked at the pathology where you have significant

00:58:05.100 burden of these neuropathological hallmarks with no decrease in cognitive function or beyond what

00:58:10.220 might be expected given that person's age. And so there's a possibility that there's some other

00:58:15.820 factor. People are now thinking about other things that may come into play here. So microglial function,

00:58:21.580 phenotype, and so microglia, the main immune cells of the brain, these are now being increasingly

00:58:26.220 interrogated in Alzheimer's disease and dementia. Other processes might be important. So lysosomal

00:58:33.340 function is potentially important as well, which basically processes proteins for breakdown and, you know,

00:58:38.540 that gets impaired and that may trigger some of the accumulation of some of these things.

00:58:42.780 But the alternative is to maybe think that some of these hallmarks may be epiphenomenal. They just

00:58:49.180 accumulate in the face of neuronal stressors. I'm speaking about this largely or pretty much

00:58:56.540 only in the case of late onset Alzheimer's disease. Your brain is going to get exposed to a bunch of

00:59:02.060 things, be that decreasing metabolic function, be that smoking vascular disease. And then as neurons

00:59:09.500 get stressed, they start to secrete some of these proteins or accumulate some of them. And there are some

00:59:15.100 lines of evidence that suggest that some of it's actually a beneficial response. So amyloid beta has

00:59:20.300 some antimicrobial and metal chelating effects. So it may be this actually a response to a stress

00:59:25.660 that's then supposed to be protective. You get to a point, and they've shown this very clearly in animal

00:59:31.260 models, where if you can force amyloid plaques to accumulate in large numbers, where they do start to

00:59:38.140 become damaging in their own right. But up to that point, it may be more of an epiphenomenon or a

00:59:43.500 response to a neuronal stress rather than this core sort of underlying pathological process.

00:59:48.300 Yeah, there's so much there to think about, right? So on the one hand,

00:59:52.860 you could say, well, look, is amyloid beta necessary, but not sufficient for amyloid

00:59:57.580 plaque? So the classic example, that would be ApoB. ApoB is necessary, but not sufficient for

01:00:02.620 atherosclerosis. You have to have it, but by itself, it's not enough. You also have to have

01:00:08.300 inflammation, endothelial damage, or dysfunction for the ApoB particle to get in there and cause

01:00:14.060 damage. But ApoB is causally related, and that's why reducing ApoB reduces events. It's necessary,

01:00:21.100 but not sufficient. If that were true of amyloid, could it explain part of the observation that lots

01:00:27.180 of people have amyloid without symptoms? Because it's only necessary, but by itself,

01:00:30.940 it's not sufficient. But that wouldn't necessarily imply causality, because to have causality,

01:00:37.820 you would have to say, removing amyloid, because it is necessary, though not sufficient,

01:00:44.220 removes the disease. And as of yet, to your point, we don't really have evidence of amyloid-reducing

01:00:51.580 strategies working. Now, the flip side to that, which I know people will argue, is it might be

01:00:57.100 because those, and I don't know where I stand on this. I'm very confused by this. It might be that,

01:01:01.500 well, we're applying those therapies too late. And if you applied amyloid-reducing therapies

01:01:08.700 earlier, if we knew how to target people long before they were at the doorstop of MCI,

01:01:15.580 maybe we could do something about it. In this sense, it's sort of like saying, well, lowering ApoB

01:01:21.180 a week or a month or even a year before someone has an MI probably isn't going to help. But doing

01:01:27.340 so 10 years before will. So how do you think about the causality and the necessary but not sufficient

01:01:34.140 argument, and also this temporal argument of, if in fact ApoB is playing a role, we may be way outside

01:01:42.380 of our window to do anything about it. I'm open to that idea still, though I am skeptical based on

01:01:51.180 the range of animal and human evidence so far. It's possible that amyloid beta is necessary,

01:01:58.700 but if it is necessary, we don't have evidence that it is sufficient to cause dementing processes.

01:02:05.660 There's been some recent work in this area that I feel supports the idea that maybe

01:02:11.820 it may not even be necessary or sufficient, which is this process called panthos,

01:02:16.940 pathological anthos, which is pathological flower. This was published by Ralph Nixon's group

01:02:22.380 out of NYU last year. And what they showed was that in a specific knockout model in mice,

01:02:30.300 when we traditionally think about amyloid plaque accumulating, we think about it accumulating

01:02:34.060 outside the neuron. And it's sort of like this whole bunch of protein that's kind of just

01:02:38.700 sticking together, and eventually that is injuring the neurons around. It's being secreted,

01:02:43.980 it's accumulating, aggregating, and then it's causing damage that way.

01:02:48.220 What they show with some really nice techniques is that what's actually happening, at least in this

01:02:52.700 model, is that amyloid is accumulating inside the neuron. It is aggregation inside failing lysosomes,

01:03:00.940 which is supposed to process proteins. And they aggregate within the neuron, which then lyses,

01:03:05.020 the cell dies, it disappears and you leave a plaque in its place. It's sort of like burst the neuron

01:03:10.620 open. And so in that setting, there's a nice quote by Nixon that says that if you're trying to remove

01:03:19.340 amyloid, it's the same thing to try and treat Alzheimer's disease. It's the same thing as trying to

01:03:25.100 revive somebody from the dead by removing their tombstone. That's essentially what it is. The plaque

01:03:31.740 is the tombstone of a previous neuron that failed its ability to process protein and then left this

01:03:38.060 in its place. There's still a lot of work to be done to say is this what's actually happening

01:03:43.100 in humans, which haven't been sort of genetically manipulated as we do with a lot of mouse models.

01:03:48.140 But that's a very stark concept, right? That's a very stark concept because it implies that amyloid

01:03:54.700 is an absolute marker for something horrible. Just as walking through a town and seeing lots of

01:04:02.140 tombstones tells you if they occurred over a short enough period of time, something bad happened in

01:04:06.940 this town. But removing the tombstones does nothing to erase what just happened in that town. What is

01:04:13.260 the critical response to their model? So the people that would be on the amyloid is at a minimum 0.99

01:04:20.460 necessary, maybe even sufficient. How would they be critical of Nixon's work?

01:04:26.540 So I think you could say that even if that's the case, even if this is intracellular accumulation,

01:04:34.380 all you've done is you've moved the site of initial accumulation. If it does hold in humans,

01:04:39.020 and maybe that would be the first major criticism is show that this happens in humans. And I think that

01:04:43.100 they're working on some pathological samples to see whether this is the case. You could say,

01:04:48.300 well, okay, first show this is how it works in humans. And then you could say, well, even if

01:04:53.500 that's the case, all you've done is you've moved the amyloid from outside the cell to inside the cell,

01:04:57.900 where it first accumulates. That doesn't then stop it, you know, causing damage, being a primary

01:05:01.660 causative factor, because, you know, once that first neuron itself and then the structures around

01:05:06.540 it could still be damaged or, you know, because then you might get microglial activation, right? You get

01:05:11.100 this immune response, right? That can then trigger some of the downstream processes. So I think there's still a

01:05:14.940 possibility that even if that's the process as it happens, you could still say, oh, well,

01:05:20.380 everything else is still the same. So let's talk about some of the other ideas, Tommy, and

01:05:25.260 we can talk about some of them broadly, but I'd like to also hear if you were Alzheimer's czar for

01:05:30.620 a day and it was your job to allocate funding for both prevention and treatment. If you were going to

01:05:38.620 allocate those types of dollars, presumably you'd want to have a strong sense of where to put those

01:05:43.980 resources. But what are some of the other theories, right? Sort of vascular, metabolic,

01:05:49.420 obviously there's a genetic component. So just tell me where you think it sort of shakes down and

01:05:53.740 then we can go from there as to what the implication is from where maybe we ought to be spending our

01:05:58.460 resources. We've gotten this far. I've talked a lot about Alzheimer's diseases. It's important for

01:06:03.740 me to mention that a lot of the work and thought I've done in this arena is not on my own. You know,

01:06:09.740 I primarily have worked a lot with Dr. Josh Turknit, who is a neurologist, which I am not,

01:06:14.380 and obviously has a lot of sort of front-facing experience in this. And so a lot of these ideas,

01:06:19.420 as they come together, we wrote a paper about the demand model. That's where some of this

01:06:23.260 discussion comes from, but it's very much a collaboration with him and others. So I won't

01:06:27.260 pretend that I suddenly figured anything out by myself because I certainly didn't,

01:06:31.340 nor can I say that I figured it out or else I'd be getting an invite from the King of Sweden

01:06:35.500 sometime to go and meet him. But when we look at late-odset Alzheimer's disease,

01:06:43.260 I think we've gotten to the point where there's such a broad number of environmental and lifestyle-based

01:06:51.100 risk factors that seem to be critically important. That's really where I would focus my efforts.

01:06:57.660 And you can definitely say that even if the brain is responding to neurological stresses or

01:07:07.900 the absence of expected inputs with the accumulation of certain pathological hallmarks,

01:07:13.740 there's still some upstream process that's driving that. And that's where I would want

01:07:17.740 to focus. And I think our best evidence so far is in factors around diet, lifestyle, and peripheral

01:07:26.060 health, you know, cardiovascular health being an important one. There was a Lancet Commission

01:07:31.500 report in 2020 that looked at dementia broadly and tried to estimate what amount of dementia,

01:07:37.980 what proportion of dementia would be preventable. And they estimated that 40 percent of dementia

01:07:44.300 is preventable based on population attributable risk and a bunch of different risk factors,

01:07:49.260 looking at physical activity, body composition, diet quality, smoking, hearing loss,

01:07:53.820 which I think of as a cognitive demand, which has lost educational status.

01:07:59.340 And my guess is that this is actually an underestimate because they didn't include sleep.

01:08:05.100 They didn't include nutrient status, particularly homocysteine status, which other individuals have

01:08:10.700 suggested has a population attributable risk of around 20 percent for late onset Alzheimer's disease.

01:08:16.700 And then they also, you know, when you do these kind of assessments, right,

01:08:20.300 and you've written about this recently, when you do a population attributable risk or population

01:08:23.820 attributable fraction, you say, if we remove this entirely, what proportion of that disease would

01:08:28.460 disappear? And we have a mounting body of evidence that says that these risk factors interact.

01:08:35.900 So when you do a population attributable risk, you say, this is a linear effect, it's additive,

01:08:41.660 this on top of this on top of this. But in reality, we know that they're actually,

01:08:45.260 they're actually interactive.

01:08:46.300 Yeah, they're accretive. Yeah. So I think if you took into account other risk factors and

01:08:51.660 more complex interactions between them, my guess is that the majority is probably preventable by

01:08:56.860 focusing on some of those lifestyle and environmental factors.

01:08:59.980 Can you say more about the homocysteine one? We definitely manage homocysteine very aggressively.

01:09:06.300 So we're very liberal with our use of methylated B vitamins to keep homocysteine down. We typically

01:09:12.060 target eight or nine as the upper limit we want to see, even though the lab reference range says up

01:09:19.180 to 13 or 14 is normal. Do you have thoughts on that? And I guess also tell me about the mechanisms.

01:09:25.340 Our concern is mostly through cardiovascular, by the way, because I think there we have pretty good

01:09:29.900 evidence that homocysteine impairs the clearance of two molecules, SDMA and ADMA that impair nitric

01:09:37.100 oxide synthase, which obviously has an important role in the cardiovascular endothelial world.

01:09:43.180 Perhaps it does in the brain as well. I'm just unaware.

01:09:46.220 So a lot of the work in this arena, I think we have to be indebted to somebody called David Smith,

01:09:51.580 Professor David Smith, who's former chair and head of the Department of Pharmacology at the University of

01:09:55.580 Oxford, who's done a huge amount of work on homocysteine and cognitive decline. And actually,

01:10:00.700 this dementia charity in the UK that I'm on the advisory board of, he is the chair of their

01:10:05.740 scientific advisory board. He has done a number of interventional studies looking at this. The main

01:10:13.100 one is the Vitacog study, where they randomised individuals with elevated homocysteine to a B vitamin

01:10:20.780 supplement. It was a B12 and folic acid. And then looked at rate of cognitive decline and rate of

01:10:26.540 brain atrophy. And they showed that if you could reduce homocysteine, and the greatest risk is with

01:10:33.500 those with a homocysteine above 13. There's also an elevated risk in those with a homocysteine above 11.

01:10:38.780 So the cutoff that I use for cognitive decline that's based on hard evidence in the clinical

01:10:44.140 literature is 11. And you can slow cognitive decline and brain atrophy if you decrease homocysteine.

01:10:50.060 Beyond that point. There's a number of potential reasons for this. The first is there may be a

01:10:56.780 direct mechanistic effect related to the neuropathological hallmarks. So homocysteine

01:11:01.660 seems to activate CDK5, which phosphorylates tau, and then also inhibit phosphatase 2A, which

01:11:09.020 dephosphorylates tau. So it may contribute to the accumulation of tau tangles, hyperphosphorylated tau,

01:11:15.500 in neurons. But then sort of more broadly and where I think the majority of the action is,

01:11:21.260 and sort of excluding cardiovascular, is the importance of the methylation cycle in creating

01:11:28.220 functional neuronal membranes. And the reason why I think this is because of evidence of how homocysteine

01:11:35.820 level or B vitamin supplementation interacts with omega-3 status. We know that if you want to try and

01:11:42.540 put particularly a molecule of DHA into a lipid membrane, you need to attach it, or you need to

01:11:48.620 create a phospholipid. And that process of creating phosphatidylcholine, say, which then attaches to your

01:11:56.140 DHA so it can sit in your membrane, it's very methyl intensive, right? So it requires, there are several

01:12:01.660 methylation steps that require that to happen. When you look at, there are actually sub analyses of several

01:12:08.700 randomized clinical trials that show, so in the Vitacog study, what they showed was that

01:12:13.260 the rate of brain atrophy was only slowed in those who had in the highest tertial of omega-3 status.

01:12:19.740 There was a study in the Netherlands, the B-proof study, where they also showed that

01:12:26.700 the benefit was greatest in those who had the highest levels of DHA. And then there's also the

01:12:31.260 omega-AD study where they supplemented with EPA and DHA, but then they saw benefit in those who had the

01:12:37.100 lowest levels of homocysteine. So there seems to be an interaction between B-vitamin status and omega-3

01:12:43.260 status in terms of cognitive decline and brain atrophy. And the best way to think about that is

01:12:49.260 probably both are required in order for DHA to be inserted in a functional way into a neuronal membrane.

01:12:55.580 Let's pause there for one second, Tommy, because I want to make sure people understand what it is

01:13:00.380 you're saying here, which is if you lower your homocysteine from 13 or 14 to 9 or 10,

01:13:09.660 all the while maybe taking some DHA, and we haven't talked about doses, but what are we thinking,

01:13:15.980 one to two grams of DHA? One to two grams a day, yeah.

01:13:20.700 Doing those two things together, which don't require any medical care, I mean,

01:13:25.420 I've never seen a person who with the right amount of methylfolate, methyl B12, sometimes some B6 is

01:13:32.780 needed. You can pretty much always get your homocysteine into that zone, coupled with a high

01:13:37.820 quality DHEA. No affiliation with any of these companies, but I like Carlson's and Nordic Naturals

01:13:43.740 as the two. What would you say is the risk reduction? Would you say that that's a 20% risk reduction

01:13:51.100 in all-cause dementia, just doing something like that? That's what others have suggested.

01:13:59.340 They've calculated population attributable risks, and it's actually, it's more than that.

01:14:03.020 Yeah, that was just the homocysteine. That was just the homocysteine. Another 20%

01:14:06.220 has been attributed to poor omega-3 status. So this is kind of frustrating because I know

01:14:11.580 that there are literally hundreds of thousands of people listening to us speak right now whose

01:14:17.500 homocysteine is elevated, whose DHEA is low, and they are unaware and their doctors are unaware of

01:14:26.220 everything you just said. And there's no drug, including a $30,000 a month drug,

01:14:33.900 that is going to come close to that level of prevention. So why do you think there is such a

01:14:40.860 disconnect in how we think about prevention? And we haven't even talked about the obvious stuff.

01:14:46.140 Like I'm just going to leave the obvious stuff aside. You should sleep. Sleep is important.

01:14:50.140 Exercise is important. Not having type 2 diabetes is important. Controlling blood pressure is

01:14:54.380 important. Not smoking is important. We're going to take those off the table. They're enormously

01:14:58.300 important, but they're so obvious we're not going to talk about them. But the homocysteine thing is not

01:15:03.180 that obvious, and yet a 20% risk reduction when you're starting with an absolute risk that's as high as

01:15:09.260 AD, that's like having a winning lottery ticket in your pocket and just not knowing it.

01:15:16.620 I can't explain why this is not better understood. When I've talked to physicians about this,

01:15:23.900 or I've had old age psychiatrists who are like, I'm immediately going to start measuring homocysteine

01:15:29.340 and supplementing B vitamins, you know, plus or minus omega threes. And I know David Smith

01:15:36.860 a little, particularly through my work, sort of through this charity. And I think he's gotten to

01:15:42.620 the point where he's incredibly frustrated. I don't want to speak for him. But basically, this pretty

01:15:48.860 significant body of randomized clinical trials showing these improved outcomes has not been

01:15:56.940 incorporated into prevention guidelines. It's not been incorporated into things like the Lancet

01:16:01.980 Commission report, which looked at population attributable risk. But based on the information

01:16:07.580 we have, and this is cheap, easy to do, you know, it's easy to measure homocysteine, it's easy to measure

01:16:12.940 an omega-3 index, if you need to, or, you know, assess the DHA intake in your diet. I think

01:16:18.700 there's, I can't say for certain, but I think there's massive potential benefit, you know,

01:16:23.180 even if it's not 40% with the combination, right? Huge potential benefit here, again,

01:16:28.140 with high-quality evidence to support it. Do you think there's just a bias in the medical

01:16:32.540 community against non-pharmacologic interventions? And I mean, look, there are few people that are

01:16:38.220 going to stand here and be more critical of the supplement industry than I am. I really think

01:16:43.020 it is a filthy industry, especially in the United States. I think it's disgusting everywhere,

01:16:48.540 but in this country, it's especially disgusting. And the total lack of regulation, the complete

01:16:54.860 predatory nature of it, and the total lack of quality control means that on average, it's filthy.

01:17:02.220 However, that doesn't mean every supplement is a bad idea. And I probably take a dozen supplements a

01:17:10.940 day or, you know, at least nine or 10, if I were to really sit there and tally them up, including,

01:17:14.940 obviously, methylfolate, methyl B12, B6, EPA, and DHA. So right off the bat, there's five. So I'd

01:17:21.500 probably take another five. But I can't help but think that there's a systemic bias in the medical

01:17:27.500 community against supplements. And some of that, I think, is probably well-founded, because I know how

01:17:34.540 frustrated, if I'm just going to be completely honest, I get when patients show up in the practice

01:17:38.700 and they have a list of 40 supplements. And I can tell just looking at them, because I've done this

01:17:43.020 exercise 1000 times, that 37 of them are garbage. But three of them are worth it. And I don't know 0.98

01:17:49.980 what the answer is. I don't know what it takes to bridge that gap. But I guess this is just an

01:17:55.740 example of where you might just have to bypass your doctor and sort of say, look, I need to know what my

01:18:00.540 homocysteine level is. I need to know what my EPA, DHA levels are. And I can fix this on my own.

01:18:06.700 And by the way, since I'm in the process of telling people what supplements I like,

01:18:10.220 especially given that I have no affiliation, I'm very happy to do this. So I prefer the Gero 0.97

01:18:15.020 supplement for methylfolate and methyl B12. I just find Gero to be a very high quality supplement in

01:18:21.420 general. I'm sure there are many others out there. I also like the Pure Encapsulations B6. So those are

01:18:28.620 kind of the variants I'm using. Do you have any supplements that you've tested or found to be

01:18:33.660 particularly trustworthy as far as manufacturers go?

01:18:37.900 For a lot of the supplements that I take, I like Thorne, just because I know they have a high,

01:18:43.340 a very stringent regulatory process. And there's multiple points in the process where they test

01:18:47.900 for impurities or contaminants. They also have a subset. So say if I'm working with a tested athlete,

01:18:53.260 that you know, if you recommend a supplement, it has to be NSF for sports certified or similar. So

01:18:58.060 you know, there's no banned products in there and they have a line like that. Pure Encapsulations,

01:19:02.220 I think is good. Momentus do some good supplements. I think part of the problem here is that,

01:19:09.740 I mean, it depends on the medical system that you're working in, right? If you're in the UK,

01:19:13.180 you're in a nationalized healthcare system. If you're over here in the US, obviously could be

01:19:16.380 several different types of system. You need to be able to prescribe it. So that needs to be built

01:19:20.940 into a system. And so in the UK, you can prescribe things like vitamin D.

01:19:23.580 In the UK, you need to have a physician prescribe those supplements to you?

01:19:29.580 No, but if you're a doctor and you want your patient to take it, then it's very hard. There's

01:19:36.220 a big gray area where you then start recommending your patient go to the pharmacy to buy a supplement.

01:19:43.180 I think part of it is this gray area, whereas you're protected if it's in some respects, if it's,

01:19:48.220 there's this indication that the National Institutes of Clinical Excellence, NICE have said,

01:19:53.100 this indication, you can prescribe it or, you know, it's available, right? There's kind of

01:19:56.780 guidelines around it. And so I think that's missing. That's part of it. But even other

01:20:01.820 physicians. So I work with a bunch of physicians who are interested in lifestyle medicine.

01:20:07.260 In lifestyle medicine, I don't want to tire multiple people with the same,

01:20:10.860 everybody with the same brush, but they're very against supplements. They're like,

01:20:15.180 lifestyle medicine doesn't include supplements. We don't do that. And anybody who does recommend

01:20:18.940 supplements is sort of not allowed in the club, which is hugely problematic because,

01:20:24.620 again, you have good evidence to say that there's benefit there. So some of it may be regulatory or,

01:20:30.860 you know, the system and how you can get your patient these things or whether you'll get in

01:20:36.620 trouble if you start getting homocysteine tests on everybody, which they can be tricky to do,

01:20:40.620 right? They have to be processed quickly. It's a slightly more expensive blood test compared to some

01:20:43.980 other blood tests. And then there's also there, like you said, I think there's some bias in there

01:20:48.060 in terms of whether people think supplements are allowed or not.

01:20:52.620 It's really frustrating. All right. Let's move on from the topic that is disappointing to me,

01:20:58.540 which is one, having a winning lottery ticket in your pocket and not playing it. And two,

01:21:04.380 being such a Puritan in one's view that you wouldn't think everything is on the table,

01:21:09.500 right? I mean, my view is, look, Alzheimer's disease and dementia in general, very, very

01:21:15.580 complicated, formidable opponent. Ideally, have as many things as possible in your quiver.

01:21:23.580 You should have a jab. You should have a right cross. You should have a left hook. You should

01:21:26.700 have an uppercut. You should have all of the above to somehow suggest we only want lifestyle or we only

01:21:31.900 want supplements or we only want drugs. I've never understood that rationale, but you're right.

01:21:36.940 Right. It permeates into cardiovascular medicine. It permeates into cancer. It permeates into

01:21:41.420 everything. And it's just created a bunch of silos of people who each have probably some truth and

01:21:48.380 some expertise, but are equally limited by their blind spots. So let's go back to the pathology again,

01:21:54.140 just really quickly. What is sort of your unifying theory on these things? So whether it is the

01:22:00.220 the homocysteine, the limitation of omega-3 marine-based fatty acids, we didn't even get into

01:22:07.820 glucose hypometabolism, but it's something we've discussed a lot on the podcast. I think people are

01:22:11.420 very familiar with hypometabolism, whether we talk about low-grade ischemia and the microvascular

01:22:17.180 disease. Do you think it's all working its way through a final common pathway of neuronal damage,

01:22:22.300 which leaves the tombstone in its wake? And that's the only thing that's basically common to them is that

01:22:27.100 they leave the same tombstone? Yes. That essentially summarizes my

01:22:31.260 current thought process. And so again, working with others, I think, including David Smith,

01:22:38.300 Josh Turk and Yintai Yu, I mentioned earlier, he did this big meta-analysis on modifiable factors,

01:22:43.900 risk factors for cognitive decline, and trying to build this systems approach to cognitive decline or

01:22:51.740 late onset Alzheimer's disease. And so then my current thought process is that all of these things

01:22:58.620 are necessary, healthy vascular supply, some kind of metabolic substrate that is taken up and available

01:23:04.700 to neurons in the brain, that you have the nutrients required to build a quality structure, that you have

01:23:12.460 the absence of things that may impair some repair process. You mentioned a bunch of them, which are

01:23:17.900 clear risk factors like smoking. And then, you know, to kind of tie everything together, you require

01:23:25.260 demand on the system that creates a stimulus for adaptation. Those things are required in order to

01:23:32.940 respond to that stimulus. And then you have a period of recovery and adaptation that allows for

01:23:37.980 consolidation and plasticity. And any individual may have an issue in one or more of those areas,

01:23:47.500 but then what ties them together is the tombstone that they leave in their wake. But the exact way

01:23:52.540 that that looks, the exact number of tombstones, say, or how those tombstones look, or, you know,

01:23:56.620 what else is going on is probably an expression of things like genetics and other factors, which then

01:24:02.140 explain some of that variability. But that's kind of, yes, it's maybe the final common pathway,

01:24:07.820 but everything that's important is happening upstream of that.

01:24:10.700 Before we leave dementia, I'd like to talk a little bit about one of the strongest associations

01:24:15.740 I've ever seen with mitigating risk, which is strength. So this is often demonstrated with

01:24:23.100 simple measurable things like grip strength, but I don't think it's that the strength of your grip

01:24:27.900 and the ability to open a jar is particularly important. It's just that very high levels of grip

01:24:33.020 strength, very low levels of grip strength, variability here is a great proxy for overall strength.

01:24:37.980 People with a very strong grip, they're able to carry really heavy things and that requires

01:24:43.340 strength all the way up and down their chain. Why do you think strength has such an important

01:24:50.220 bearing on both the avoidance of dementia, so from an incident standpoint, and also survivability?

01:24:56.940 So just to give people some numbers, when you compare the top 10% or so of people from a strength

01:25:03.180 perspective to the bottom 10%, it's about a 70% reduction in both incidence and mortality associated

01:25:11.580 with all-cause dementia. When you start to think of things that we have some control over, this also

01:25:17.740 rises to the top of the list. I'm not suggesting that someone at the bottom can be at the top within

01:25:22.460 a year, but if you think about this over the course of your life, this is something we can all aspire to.

01:25:27.180 What do you think, from a physiologic standpoint, explains such a stark relationship?

01:25:34.380 So my experience of talking about muscle mass and muscle strength with hard outcomes, say dementia,

01:25:42.700 or all-cause mortality, usually the response is, all right, calm down, bro. Clearly you like to lift

01:25:48.700 weights and therefore you think that that's the answer to everything. And people who are healthier

01:25:55.100 are stronger and that's just the confounding factor. But I think we have good evidence that

01:26:00.620 that's not the case. The main one being that you can take individuals in their 70s,

01:26:05.420 you can put them on a very basic resistance training program, and you can see improvements

01:26:10.220 both in white matter connectivity, you can do an MRI scan, or you can test them on various tests of

01:26:15.580 cognitive function and you see improvements as a result of this training program. So then,

01:26:21.900 if you want to think about potential reasons why, the simplest is that some kind of novel

01:26:28.940 movement is a direct neuromuscular stimulus. You are stimulating the brain to create new connections,

01:26:35.260 driving plasticity, because the recruiting of motor fibers, that motor skill is in itself a stimulus,

01:26:41.820 a cognitive stimulus. But then we also know that the muscle that you have and the amount that you move it,

01:26:48.700 it's an important glucose sink. So if you think about blood sugar regulation, and we know that

01:26:53.020 pre-diabetes and type 2 diabetes, like you said, are significant risk factors for cognitive decline

01:26:57.900 and dementia. So it may be increasing glucose flux that helps regulate blood sugar. We also, you know,

01:27:04.540 if you're moving your muscles, there's this still exploding area looking at myokines,

01:27:10.620 so things released by muscle tissue that may support brain function, IGF-1,

01:27:15.340 VEGF, BDNF, brain-derived neurotrophic factor, that may support neuronal function,

01:27:19.740 these sort of survival factors to keep neurons around. We know that exercise is also, through a

01:27:25.260 hormetic process, anti-inflammatory. So chronic inflammatory conditions we kind of mentioned

01:27:30.380 earlier, but can be associated with increased risk of dementia, and you're decreasing systemic

01:27:35.420 inflammation through physical activity. So that kind of group of things, I think all of those are

01:27:42.540 playing a role. I don't know exactly which is most important, if any. Frankly, I'm not sure I really

01:27:48.380 care that much, because I know that it's a very important intervention. So, and again, it's documented

01:27:54.140 that it can actually work, and you can implement it pretty much at any stage. So one of the reasons

01:27:59.660 why it is so beneficial is because it has these multiple pleiotropic effects, and they're probably

01:28:03.980 at least additive, if not synergistic, as well as, yes, there will be a certain amount which says that

01:28:09.500 the healthier you are, the stronger you are. So there's a little bit of that, but then

01:28:12.860 there's a whole bunch of things on top of that. When I met with similar resistance,

01:28:17.340 no pun intended, to the idea of the importance of strength and muscle mass, I usually respond

01:28:22.140 with something which is, look, even if strength training, meaning have a high degree of strength

01:28:26.940 and muscle mass and cardiorespiratory fitness, even if all of those things didn't add a single day

01:28:31.980 to your life, in fact, even if they shortened your life by six months, they would still more than be

01:28:37.420 worth it in terms of the quality of your life, especially in that final decade, which is

01:28:42.380 something that most people, unfortunately, don't want to pay attention to until they're in that

01:28:47.980 marginal decade, as I call it. And you realize that at that point, having poor movement, being in pain,

01:28:58.220 having low strength, limits your capacity for doing just about everything that people would find

01:29:04.620 pleasure in, whether it be playing with their kids or simply going for a walk or carrying out

01:29:09.500 any basic activity of daily living at the extreme level. So it's a fitting way to end our discussion

01:29:15.180 on this before we transition. So the other thing I want to talk about, Tommy, is another area where

01:29:19.340 you have a great amount of expertise and that is around head injury. About 10 years ago, I had a friend

01:29:25.660 that suffered a very significant head injury riding a bike down the street and a jogger bolted out

01:29:32.780 between some parked cars, didn't notice him riding, and they hit head to head. The jogger got the worse

01:29:40.060 of it because he was wearing a helmet, but nevertheless, they were both devastated by this. He was going about

01:29:45.580 40 kilometers an hour. She sustained multiple fractures to her head, but he sustained a concussion that was

01:29:51.900 so bad that basically he wasn't himself for about two years. He's more or less himself now. Can you

01:29:58.940 explain what a concussion is? Even this, the idea of what concussion is, is quite hotly debated. But

01:30:08.220 in general, you would classify a concussion as a mild traumatic brain injury. With more severe traumatic

01:30:16.380 brain injuries, you might think of like complete open skull fractures and direct penetrating

01:30:21.500 trauma to the brain, things like that. So this is, the skull remains intact, but there has been

01:30:26.940 some transmission of force or a wave, a blast wave, say if it's a blast injury that's been transferred

01:30:34.540 to the brain. Then at some level, in order to have symptoms of a concussion, you have some kind of

01:30:42.540 disturbance of neuronal function. And that can either be because of abrupt loss. So there are some

01:30:49.180 significant head impacts in particular where you can get shearing of axons, direct axonal injury of

01:30:57.580 the neurons, and then that cell is essentially lost as you've sort of ripped it apart. But even more,

01:31:04.620 you know, milder impacts may cause disturbances that include like a neuron firing when it shouldn't.

01:31:11.580 This can then sort of create this pattern of activation, again, that's not expected or in an area of

01:31:18.140 the brain where it wouldn't normally occur in a way that it wouldn't normally occur. And then this can

01:31:22.220 then cause these downstream processes within cells that can cause mitochondrial damage, swelling, you

01:31:29.100 then might see the accumulation of certain pathological proteins. So tau is just like you

01:31:34.620 see an Alzheimer's dementia, it's also a response to direct neuronal injury in a concussion. I work

01:31:41.580 with a neurosurgeon, whose definition of concussion is any impact or force to the brain that causes the

01:31:48.620 disturbance of function of one neuron. Unfortunately, that's not something that we can measure, because

01:31:53.500 you probably need multiple, or you know, large sections of the brain to have aberrant function for

01:31:57.980 you to be able to actually measure it or detect it. But those are the various processes that are going on

01:32:03.260 when you get a head impact. So it could be as quote unquote mild as just a headache for a few days,

01:32:09.820 following a head trauma, or it could be in the case of this friend of mine for a couple of years,

01:32:15.420 they had a lot of photosensitivity, they had a lot of auditory sensitivity, they had difficulty,

01:32:20.860 you know, processing things, they were much more irritable. How common are those types of more

01:32:25.820 severe symptoms than just a headache for a few days after a hit to the head?

01:32:29.660 So, in reality, it's quite difficult to say how common things are, because

01:32:38.300 millions of concussions happen every year in the US alone, most of which probably go unreported.

01:32:46.140 And so it's only when you see more significant symptoms, or something's happened in front of a,

01:32:51.260 so you're playing sport, it's happened in front of a doctor, you know, you get a formal assessment.

01:32:54.540 And, you know, the downstream effects are, you know, are varied, you know, can be around verbal

01:33:00.860 effects, photosensitivity, noise sensitivity, effects of memory, focus, reaction time, you know,

01:33:07.900 depending on how you're measuring things. And probably some of it is exactly which area of the

01:33:13.180 brain was impacted. And then there's a systemic response aspect as well. So you can only really do

01:33:19.980 this in animal studies, but part of the process of the sort of the disease process after traumatic

01:33:25.100 brain injury is a systemic immune response that seems to sort of contribute to some of the symptoms.

01:33:29.980 So how much of an inflammatory response you get, how much of a fever do you get,

01:33:34.540 that can then cause some issues throughout the brain as well.

01:33:37.820 So what do we know about the short-term management of concussion and the long-term? And what is the

01:33:47.740 relationship between a concussion and a traumatic brain injury, which are terms that I think people

01:33:51.980 are hearing. And of course the term TBI in the past decade is sort of more front and center for

01:33:56.380 people. I suspect in large part due to two things, right? Probably greater understanding of TBI occurring

01:34:03.020 in the battlefield and also in at least one sport, American football, though obviously it occurs in more

01:34:07.660 than that. Yeah. So like I said, the concussion is generally considered, so all of these are TBI,

01:34:16.140 traumatic brain injuries of various severities. A concussion is sometimes also known as a mild

01:34:23.340 traumatic brain injury or MTBI. And I think this is where probably most of our focus today will be on,

01:34:30.940 because the more severe brain injuries, so say if you do have skull fractures or penetrating head trauma,

01:34:36.460 or you have significant loss of consciousness that requires intensive care, that requires specialist

01:34:43.980 neurosurgical treatment that's probably beyond what most people here will be thinking about.

01:34:49.740 But if instead you're thinking about concussions on the sports field or blast related or sub-concussive

01:34:57.180 exposures, which might happen again in sports or in the military, and there's increasing evidence that

01:35:02.540 says that sub-concussive impacts or blasts, so an example would be sniper fire, right? You essentially

01:35:09.980 have a mini explosion happening right next to your head several times a day if you're in the range

01:35:14.860 practicing. And there's some evidence that suggests that even that over time the damage can accumulate and

01:35:20.940 cause issues with cognitive function. And there you call it sub-concussive because each individual

01:35:26.380 one doesn't necessarily cause symptoms, but depending on how you define concussion you may or may not get

01:35:31.100 symptoms anyway, so there's kind of a grey area there. But each individual one would, you wouldn't

01:35:36.060 be able to detect it really, but the damage seems to accumulate over time. If you have a significant

01:35:42.540 head impact, so say on the sports field is probably again where people are most familiar with it,

01:35:47.580 there's probably two aspects that are worth separating. So there's the formal medical process

01:35:52.540 and medical assessment that you should undergo, and there are people who may, you know, in multiple

01:35:58.780 sports now, they may have some baseline cognitive testing so that when you do get a concussion they

01:36:03.340 retest you, they make sure that you get closer to your previous baseline before you're allowed to play 0.99

01:36:08.460 again. So things like the impact assessment is often used in the sort of youth and college sports.

01:36:14.460 And so there's that part. But then what I'm particularly interested in is how can we mitigate

01:36:21.500 the effects of an impact initially? Like are there supplementation of other strategies that we can do

01:36:25.740 to make the athlete or somebody who's at high risk of concussions more resilient to that impact?

01:36:32.700 And are there other things that we can do to support recovery or minimize the effect of the impact?

01:36:38.300 So with the short term post concussion management, there are two or three things that I think are

01:36:45.100 particularly important. So the first is thermoregulation, and which is basically managing

01:36:51.900 normal body temperature. My PhD was actually looking at the effects of temperature on response to brain

01:36:59.340 injury. And it's very clear that the hotter your brain gets after an injury, the worse your outcome.

01:37:06.540 Pretty much any type of brain injury, stroke, traumatic brain injury, also these neonatal and

01:37:12.700 pediatric brain injuries that I study. And we know that a lot of sports happen in heat stressed environments.

01:37:20.300 So you're on the field, you're hot already, or you're out in the sun. And experimentally, if you

01:37:27.900 heat that brain up first, which happens during exercise, and then you have an impact,

01:37:32.620 and then that brain stays hot, that seems to worsen outcomes. So getting somebody out of a heat

01:37:38.060 stressed environment, cooling them down if you need to, and you can use external ways to do that,

01:37:43.420 you can use things like Tylenol to help regulate body temperature. And that particularly becomes

01:37:47.580 important because this, at some point, a few hours later, an inflammatory response is going to kick in,

01:37:52.140 and it's very common to get fever. How long does that power or potential of that intervention last?

01:37:58.140 So if a person has a concussion, you know, at two o'clock on a Sunday afternoon,

01:38:04.780 should they be in an environment of taking Tylenol and cooling themselves off for the rest of the

01:38:09.500 day? Or is that something that you'd want to carry out for a couple of days and just continue

01:38:13.820 round-the-clock Tylenol to prophylactically ward off fever?

01:38:17.580 That's a great question. And it's something that the field probably still struggles with. But a

01:38:23.580 period of 24 to 72 hours after the initial injury is probably the most critical. The most important

01:38:30.060 thing is preventing fever. And the reason why fever causes issues is because you increase the mismatch

01:38:37.260 between metabolic rate and capacity to produce energy to match that metabolism. So if you have

01:38:42.780 mitochondrial dysfunction, you're basically increasing that gap in terms of required energy

01:38:48.300 production, which then exacerbates the injury. Any evidence for cooling, Tommy? So for example,

01:38:53.820 I mean, obviously, as you know from your work, when you're doing certain types of heart surgery,

01:38:59.660 you can actually cool the patient to 19 degrees Celsius. At least back when I was training,

01:39:05.420 that was the sweet spot. I don't know what the temperature is today. But if you were basically

01:39:10.540 doing anything where you had to cross-clamp the aorta and prevent cerebral perfusion, you would

01:39:15.340 cool the patient, literally have ice around their head. The anesthesiologist would cool the patient.

01:39:19.100 And obviously that was very protective. So is there any evidence that someone sustaining a concussion

01:39:23.820 should go beyond just getting out of the sun and taking Tylenol, but perhaps be laying down and be,

01:39:29.340 you know, bathed, or at least have their head covered in ice?

01:39:32.540 The short answer is no, there's no evidence for that. And then there's a long story to answer that

01:39:37.980 question. So in animal models, and again, this is something that I've published extensively on,

01:39:44.860 hypothermia is magic for acute brain injuries, right? If you can decrease core temperature by

01:39:50.780 three to four degrees Celsius for 24 hours to 72 hours after the injury, you get a significant

01:39:57.020 reduction in brain injury. In one specific scenario, and that's acute brain injury in babies that have

01:40:05.340 some kind of issue around birth, therapeutic hypothermia, so cooling to 33.5 degrees Celsius core

01:40:10.860 temperature for three days is the standard of care. It was brought into the resuscitation guidelines in

01:40:15.660 2010. And they do that just externally. They're not using ECMO or anything like that.

01:40:20.540 Okay. No. So some of those kids can go on ECMO, but that's because of respiratory failure.

01:40:25.020 Got it.

01:40:25.500 And so you can cool on pump. But no, this is just an external cooling with a blanket. In models of TBI,

01:40:34.140 it is similarly neuroprotective in animals. Huge bodies of work say that if you create a traumatic

01:40:39.660 brain injury in an animal model, and then cool the animal down, they get decreased injury.

01:40:44.220 Dozens of trials, billions of dollars have been spent trying to replicate this in humans, and it has not

01:40:49.420 worked. And people thought the temperature wasn't right. The duration wasn't right. You know,

01:40:54.460 obviously, in traumatic brain injury, you have a very heterogeneous population. Maybe the population

01:40:58.220 wasn't right. But in reality, there's no evidence that hypothermia works for concussions or any kind

01:41:04.620 of traumatic brain injury. But what does seem to work is maintaining normothermia. So basically,

01:41:09.420 a core temperature at or below 36.5 degrees Celsius, ideally. There are some people out there who will

01:41:16.540 sell cool caps and things. And I think I once spoke on a podcast and said what I just said.

01:41:22.620 And somebody emailed me and was like, here's my unblinded, open study of my cool cap in some

01:41:28.940 hockey players that shows that it improves recovery from concussion. There is no high quality evidence

01:41:34.140 that these things work. So particularly because externally cooling the brain, you're probably not

01:41:38.860 going to get the brain cold enough to do it. What you need to do is you would need to cool the blood

01:41:43.020 going up into the brain. So but in reality, I don't think there's much evidence there. So

01:41:47.980 focusing on preventing fever or managing fever, I think is very beneficial. But then active cooling

01:41:53.260 below that doesn't seem to add anything. And these studies, they demonstrated they were

01:41:59.420 actually cooling the neck, they were demonstrating that they were achieving a significant reduction in

01:42:04.380 brain temperature. So these external devices like cooling the neck in individuals with concussion,

01:42:10.860 that hasn't been shown to be beneficial when when they've trialed hypothermia for traumatic brain

01:42:16.060 injury. Oh, I just mean, but did they demonstrate efficacy of cooling? Not that it reduced concussion,

01:42:20.860 but did they show that they can cool the brain? No. So this is another issue is that particularly with

01:42:26.860 milder external forms of cooling, how do you know the brain is actually being cooled? And so,

01:42:32.700 so in these studies, you may see like particularly they're doing external cooling on the scalp,

01:42:38.060 they'll measure scalp temperature and say, well, the scalp is colder, therefore the brain is colder.

01:42:42.220 That's not necessarily the case, right? So so no, there isn't documented evidence that these things

01:42:47.660 can directly significantly cool brain temperature. So there's two problems, right? One is we don't know

01:42:53.900 if we're even cooling the brain in these studies. And of course, therefore, we don't know if cooling the

01:42:59.740 brain would minimize or mitigate this risk. It could be the complexity of the human model.

01:43:04.220 And given our surface area, you know, infants have a surface area amount that allows for easy

01:43:11.340 external cooling. In fact, that's one of the challenges of taking care of pediatric patients

01:43:15.500 is they're very susceptible to insensible losses in the way that adults are not. So here you're using

01:43:20.620 it to your advantage. So that's very interesting. If it doesn't work, in other words, if we could be

01:43:26.860 sure that those studies are indeed cooling the brain, but it is not having an impact, what is your

01:43:32.940 explanation for the discordance between that and the pediatric and animal literature?

01:43:38.940 So there are two potential things that could conceivably remain as answer questions. So one

01:43:47.500 is that the majority of cooling studies, although this has been overcome in a lot of them, but the

01:43:53.020 majority of cooling studies are probably not cooling soon enough. So there's actually a very narrow window,

01:43:59.340 again, based on animal studies that says, in an acute brain injury, you basically need to start

01:44:05.340 cooling therapy within six hours, but probably ideally within three hours of the injury. Now,

01:44:11.020 if you're running a clinical trial, it's almost impossible, right? It's the same as doing thrombolysis

01:44:16.060 in stroke, like trying to get it as soon as possible. It's very difficult. So if you could immediately

01:44:21.020 cool, maybe that's part of the issue. The other issue, which is potentially a little bit

01:44:27.260 controversial, but I think it's increasingly being appreciated in the neonatal literature is that

01:44:34.220 when they did the original cooling trials, the control group were kept at 37 degrees Celsius and had a

01:44:42.060 lot of fever. Oh, you mean the controls were managed with Tylenol and thermoregulation to keep them at 37?

01:44:50.540 No. So originally, in those original cooling trials in neonates, the control group was hot. So if you

01:44:58.620 look at a baby's normal temperature after they're born, they don't immediately go to 37 degrees

01:45:02.940 Celsius. They may never get there. They'll slowly warm up over a day or so. But in those trials,

01:45:09.500 the control group were immediately warmed up to 37 and they also frequently had fever and they weren't

01:45:15.340 managing those fevers because that was in the era where you were worried about the kids getting cold.

01:45:20.620 Yeah. So in other words, you may have made the controls worse off and artificially created a gap.

01:45:25.820 Exactly. So in more modern cooling trials, they do targeted temperature management in the control

01:45:31.340 group. So the control group is kept at 36.5 degrees Celsius core temperature. And then compared to that,

01:45:39.500 the cooling group doesn't seem to have a benefit. So some of the effect in neonates may be due to

01:45:45.980 a worsening of outcome in the control group rather than benefit in the intervention group.

01:45:51.820 Very interesting. Okay. Another topic I want to ask you about is hyperbaric oxygen. And presumably,

01:45:59.660 this will be in the chronic phase. I mean, maybe someone's looking at this in the acute phase,

01:46:03.820 but address it as you see fit. But clearly, I see a lot of people talking about hyperbaric oxygen.

01:46:09.180 I get asked about this all the time from patients. I'm quite ambivalent about this. So I generally

01:46:15.020 think the hyperbaric oxygen industry is a racket. I think there are definitely indications for it.

01:46:20.300 There are clearly places like wound healing where I think it matters. I probably think the opportunity

01:46:26.860 cost notwithstanding, it might be reasonable to consider hyperbaric oxygen for brain injury,

01:46:32.380 but I haven't really found any compelling data. So it's possible I just haven't been looking hard

01:46:37.100 enough and obviously you're closer to it. So what's your state of the understanding of the role of that?

01:46:43.100 I think the phase is critically important. Firstly, because in the acute phase of injury,

01:46:51.340 there is evidence that hyperbaric oxygen can be detrimental. I know of some studies where they've

01:46:56.940 tried to get hyperbaric oxygen in as early as possible after traumatic brain injury, and they've had

01:47:01.820 an increase in negative side effects. I don't think those studies are published yet, but I've

01:47:05.580 certainly heard of that work. So presumably just creates more reactive oxygen species in a

01:47:10.700 hyperinflammatory environment. That's exactly it. And again, so if you have impaired function,

01:47:15.100 and then you try and shove a bunch of extra oxygen in there, you're essentially just driving oxidative

01:47:18.780 stress. And I know of some people who have actually, you know, there have been significant

01:47:23.900 injuries in like professional sports, and they've immediately gotten that person in the chamber.

01:47:27.580 That is not something that I would recommend, and I don't think there's evidence to support it.

01:47:31.660 Longer term, so in the chronic phase, I think there's a possibility of benefit. And again,

01:47:37.420 there are some data out there to suggest that those after concussion or traumatic brain injury,

01:47:41.260 they may see improvement after exposure to multiple rounds of hyperbaric oxygen. However,

01:47:47.820 there are a couple of things that just should drive some equipoise currently. One is that most of these

01:47:52.380 trials are single arm, uncontrolled studies. And it's actually quite hard to create a control group

01:48:00.780 in hyperbaric oxygen studies, because you can tell when the chamber is being pressurised. And so to try

01:48:08.380 and create a scenario where you have a sham effect that may create the same placebo is quite tricky.

01:48:13.740 The second is that I read a study where they're like, these people, they all improved, you know,

01:48:20.460 we did six months of hyperbaric oxygen, they go twice a week. And in my mind, you're just invoking

01:48:26.060 Voltaire, who said that medicine is the art of entertaining the patient while nature cures the

01:48:30.780 disease, right, that brain was going to get better anyway. Having said that, I think there are some

01:48:37.100 early data that suggests that, you know, improved cognitive function, you know, maybe you're restoring

01:48:41.980 metabolic function within the brain mitochondrial function with some periods of hyperbaric oxygen,

01:48:46.380 it probably takes several rounds. So like, a couple of months, at least 30 plus exposures

01:48:52.060 seems to be required. And again, this is sort of like, anecdotal, I've seen people's pilot data,

01:48:56.300 this is not large randomized controlled trials or anything like that. I think there is potential

01:49:00.060 benefit. How long after you're saying anytime after 72 hours, or you'd wait even longer potentially?

01:49:05.980 I'd probably wait a couple of weeks, would be my guess until we're sort of like the full initial

01:49:10.460 phase of the injury has resolved. And you're basically saying two 60 minute

01:49:15.820 sessions weekly, and they're doing this to what depth? Or what's the whatever metric of

01:49:21.420 barometric pressure? I think generally, they do two atmospheres, I think is generally what's done.

01:49:26.220 Two atmospheres, an hour at a time, twice a week, and potentially several months, yeah.

01:49:31.900 Well, it's certainly long enough for Voltaire to take over. So it's very difficult to know.

01:49:37.340 What about supplements such as creatine monohydrate and our good friends, the omega-3,

01:49:45.020 the marine omega-3 fatty acids, DHA and EPA? Do they play any role here in supplementation if a

01:49:51.660 person is not using them, i.e. adding them? Or is there any benefit for using higher doses than

01:49:57.900 people might use? We typically talk about using five grams of creatine monohydrate daily. Talked

01:50:03.820 about that in the past on the podcast, lots of benefits associated with creatine monohydrate

01:50:08.060 use, both cognitively and physically. And we talked about two or three, maybe even up to four

01:50:12.780 grams of EPA, DHA having benefit. What's the evidence for their efficacy here? And how do you

01:50:18.300 think about dosing them? So I think the doses are very similar, but I do think there's a strong case

01:50:25.420 for benefit. There's no good study that gives a bunch of people creatine before they get traumatic

01:50:32.300 brain injuries. That's one of the problems with TBIs. It's difficult to predict. Therefore,

01:50:36.380 it's hard to do good quality prophylactic studies. But in animal models, you can see that if you

01:50:42.700 supplement with creatine for several days, the optimal period may be five to seven days with a

01:50:47.740 loading type dose. So the equivalent of 0.2 grams per kilo, something like that, which would be 20

01:50:53.820 grams per day for a week. The equivalent in both rats and mice before traumatic brain injury is

01:50:59.020 significantly neuroprotective. That kind of dose and time periods, so something like 20 grams a day

01:51:05.820 for a week has been shown to significantly increase brain creatine levels by doing something like a

01:51:11.740 magnetic resonance spectroscopy. You can point an MRI scanner at the brain and you can see the levels of

01:51:16.860 creatine in it. At the same time, there have been some studies in high school football athletes that

01:51:22.540 have looked at that same measure over a football season and seen that related to the number of

01:51:29.500 impacts that the kid gets, there's a decrease in brain creatine in the dorsolateral prefrontal cortex

01:51:35.180 and the primary motor cortex. So suggesting that creatine is being used up by impacts and provides

01:51:43.100 this sort of short-term pH and energy buffer that may be protective when there's an impact.

01:51:48.700 And those kids are taking 20 grams a day throughout the season or five grams a day?

01:51:54.140 No, these are two separate lines of evidence. Oh, so that's just looking at innate or endogenous

01:51:59.500 creatine? Yeah, so they're not supplementing. It's just looking at endogenous levels of creatine

01:52:03.980 that decrease throughout the season and it's related to impacts. So we know that with standard dosing of

01:52:09.100 creatine, we can increase our brain creatine levels. And we also know that with impacts,

01:52:13.740 brain creatine seems to decrease. And one of the thought processes behind something called

01:52:18.780 second impact syndrome, which is that you have a concussion, you recover okay, you seem to be doing

01:52:24.460 okay. That first concussion or that first impact doesn't have a big effect, but the next one,

01:52:28.540 it was an outsized effect. Yeah, we saw an awful example of this in the NFL this year with the

01:52:33.100 Dolphins quarterback, right? Yeah, yeah, exactly. And one of the things that may be happening there,

01:52:38.380 some people think that at least part of that is you depleted brain creatine with that first impact,

01:52:43.580 and then there's a greater effect of the second impact. And there's probably maybe multiple things

01:52:47.180 that happen because brain choline also seems to decrease in a similar manner. So I think there's

01:52:52.700 there's a case to be made for prophylactic creatine in those who are at high risk. If you can estimate

01:52:59.100 when you're going to be at highest risk, I may do a loading period 20 grams for a week. So you know,

01:53:03.260 you can increase brain creatine. But then after impact, there are several studies that and I think

01:53:09.100 this is what you've talked about previously, where creatine may improve some of the problems that

01:53:13.500 happen after a concussion. So creatine can offset some of the cognitive deficits caused by sleep

01:53:19.820 deprivation and sleep can be an issue with those with concussion. In multiple studies, creatine

01:53:25.180 supplementation improves cognitive function, the greatest effect seems to be in those who are oldest,

01:53:29.900 but if you have some deficit in cognitive function, which may be related to creatine if

01:53:33.580 it's through a concussion, then supplementation may be beneficial. And finally, there's some evidence

01:53:39.660 that it may help with mood. Creatine has been tested in two randomized controlled trials where

01:53:44.220 they've added it on top of SSRIs in those who responded poorly or incompletely to the SSRIs and saw

01:53:50.060 improvements in depression symptoms. So I think there's multiple reasons why if you're at high risk of

01:53:55.900 a TBI or you've had a TBI, it's worth thinking about creatine supplementation.

01:53:59.340 And presumably, if you have long enough time, just being on five grams daily will get you what

01:54:02.940 the 20 gram loading dose do. Because again, to your point, if you can't predict when you're

01:54:07.260 going to have a TBI, you can't always be taking 20 and waiting. If you just took five every day,

01:54:11.900 you'd produce similar levels within the brain. Tell me about your thoughts on EPA or DHA specifically

01:54:18.380 around this. So for relates to concussion, I think DHA is probably the more important one. And it's very

01:54:27.420 similar actually to some of the processes that we thought about with cognitive decline around

01:54:31.020 neuronal structure and normal neuronal function. There was one nice study where they looked at

01:54:36.940 supplementing with different levels of DHA again throughout a high school football season. And

01:54:42.060 what they saw, they supplemented with two, four or six grams of DHA. And actually all of those groups,

01:54:48.380 with no greater benefit with a higher dose, saw a decrease in play-related circulating neurofilament light,

01:54:54.460 which is a marker of neuronal injury that you can pick up when you get a concussion. So across the

01:54:59.740 season, neurofilament light tended to increase in the athletes, but that was mitigated in the DHA

01:55:04.700 supplementation groups. Is that a biomarker you see in the blood or is that something you see on,

01:55:09.180 can you tag for it on an MRI? No, it's a biomarker you see in the blood. Interesting. Are there any

01:55:15.100 other supplements that you think should be a part of the toolkit for basically anybody? I mean,

01:55:21.660 here's the reality of it is we're all kind of at risk for TBI. You get into a car accident,

01:55:26.380 you have a TBI. You're walking down the sidewalk and someone bumps you while they're on their little

01:55:30.940 electric scooter, you can get a TBI. So it's certainly the case that different occupations,

01:55:35.980 different sports, different stages of life will have a higher versus lower risk. But it's hard for

01:55:41.820 me to imagine that if there's a low risk way to mitigate it, we shouldn't all be doing it,

01:55:46.700 given that none of us know when we're going to get hit by an aberrant softball or get into a car

01:55:52.220 accident or fall while we're skiing or whatever it is that we do. So what would you put in the

01:55:56.780 category of kind of no regret moves along with DHA and creatine, which already have so much benefit

01:56:03.420 that we've discussed in other podcasts vis-a-vis brain health. And in the case of creatine, not just

01:56:08.060 brain health, but also muscle performance. Anything else that rises to the level of you got to think about

01:56:14.380 it? Yeah. So assuming that other things are taken care of, so right, you know, you're in good

01:56:20.300 metabolic health because we know blood sugar regulation is going to be important. And you

01:56:24.700 know, your homocysteine isn't 15. That's going to be important. Then the final thing that I would say

01:56:29.740 goes in that list is choline as citicoline or CDP choline. And do you get enough of that just eating

01:56:36.220 eggs or things that are rich in choline? Yeah. So if you eat a couple of eggs a day on average,

01:56:42.380 some things with lecithin in, phosphatidylcholine, you're probably getting enough from the diet.

01:56:47.820 But post impact, I would probably take one to two grams per day. And there's some evidence that

01:56:53.340 not that choline helps with very severe traumatic brain injuries, but in survivors,

01:56:57.420 there's some improvement in neuropsychological outcomes in those who supplement with choline.

01:57:00.540 So that's another one. Those three things I think are definitely worth considering.

01:57:04.700 And tell me the format you would supplement the choline in.

01:57:07.100 Citicoline slash CDP choline, which is the same thing. But alpha GPC, which is the other form of

01:57:13.260 choline is much less studied in that context. So I would take citicoline.

01:57:17.340 Okay. And again, presumably we're going to the same brands that we're talking about,

01:57:21.900 right? Whether it be Thorne or Gero or Pure Encapsulations kind of go with a super high

01:57:26.780 quality brand. Even if you're going to pay a little bit more, it's a relatively small price to pay

01:57:30.780 to consider the alternative. Tommy, this is super interesting. Amazingly, we have not really

01:57:37.020 spent much time on F1. Maybe I'll close with one F1 related question. For the casual observer of F1,

01:57:45.580 maybe someone who knows a little bit about it, but not a lot, you get to spend a lot of time on the

01:57:49.260 inside of F1. What's something that you think would surprise the casual observer about the sport and

01:57:55.820 maybe something about the demands of the driver? What do you think would kind of endear people to

01:58:01.100 pay closer attention to F1, which is of course my secret ambition with this podcast?

01:58:06.220 So I'm not sure if this qualifies for your last request, but the thing that surprised me the most

01:58:12.700 is probably related to the first things you said. And that is how much attention is paid to the health

01:58:19.580 and performance of the driver relative to the health and performance of the car. Historically,

01:58:25.260 in F1, the car was everything. Basically, you would just bring in a driver who was good,

01:58:30.380 and that was it. You'd let them get on with it. And because of the amount of travel, the amount of

01:58:37.580 media commitments, so I mean, they probably spend half their time with media commitments, right,

01:58:41.580 rather than working on the car or working with the engineers or in the simulator, whatever. So probably

01:58:45.660 the amount of time they have to do things like that. But because of all of that, the capacity

01:58:51.180 to do a bunch of things related to their health and performance is quite small. They just don't

01:58:56.540 have that availability of cognitive or time resources to do it. So they'll spend a bunch

01:59:01.900 of time with their coach or trainer in the gym, aerobic work, all that stuff is important. But then

01:59:08.060 things on top of that, it's very variable from driver to driver how much they focus on it. And it's

01:59:13.580 also variable how much they're willing or interested in focusing on that. So what was useful to me going

01:59:21.020 into that world is, and I imagine you would be the same, right? You'd show up and be like,

01:59:24.700 I have a list of 100 things that are going to help. Here it is. And when you work with professional

01:59:31.180 or high level endurance athletes, right, solo athletes, they're all, it's all their own

01:59:34.940 performance. And they're usually type A's, they will do it, they will do all 100 things,

01:59:39.900 no questions asked. In that setting, you have to be really certain that the thing that you're asking

01:59:46.060 that driver to do is going to be beneficial. So, you know, even if you don't have hundreds of

01:59:51.340 randomized controlled trials, you have to be, you're really working with that sort of this idea

01:59:55.580 of positive asymmetry. So like, very little risk, very high potential benefit, whatever it is. And

02:00:02.220 you're also selecting that one thing, instead of the 99 other things on your list. So it really forces

02:00:09.820 you to focus on what's going to be the most impactful thing that's actually likely to be able to make it

02:00:15.500 into sort of the driver's processes, and have a strong feeling that that's the thing that you should

02:00:20.940 really be focusing on. Yeah. The opportunity cost is so great. And there's so many other demands,

02:00:26.780 as you said. I mean, understanding the balance of the car, understanding, because I think what most

02:00:31.340 people probably also don't understand that we take for granted as sort of diehards is you've got 23

02:00:37.420 races in the year. Every race, the car has to be a little bit different. Every circuit is different.

02:00:43.740 So a street circuit like Monaco has pretty much nothing in common with a big track circuit like

02:00:50.700 Monza. And a circuit that has lots of high speed corners without many low speed corners is totally

02:00:56.780 different than the reverse. And circuits with long straightaways, where you want very low drag,

02:01:04.380 very different from circuits that are short and fast. And each of those changes everything about the

02:01:11.020 dynamics of the car, changes everything about how temperature gets into the tires, stays in the

02:01:16.220 tires, degrades, track surface, all these things. So the minute they finish that race on Sunday,

02:01:21.180 they've got two weeks, sometimes one week to completely change the car, to optimize it for

02:01:27.340 what they think it's going to do and behave the next week. And the driver has to be ready to do that.

02:01:32.140 And yeah, and then someone like you comes in and says, all right, we've only got this much time to

02:01:38.300 get you physically ready. And of course, the jet lag demands are sort of insane, right? I mean,

02:01:41.820 it's really hard to fathom how much they are crisscrossing the globe. Actually, there's a great

02:01:46.540 little video I saw on Instagram that showed the flight path of the season. And it did it to the

02:01:53.660 awesome like F1 theme song. It's just so perfectly done. Without naming any names, what intervention or

02:02:02.300 advice that you've given a driver are you most proud of in terms of the impact it's had on his performance?

02:02:07.980 Something that that I really enjoyed, or I thought it was really impactful recently was

02:02:14.940 helping a driver who was focusing on time off the start line. And of course,

02:02:21.340 in general, I'm not directly interacting with the driver, the driver has a coach. And a lot of

02:02:25.900 the interactions is me and the coach and tinkering with things the coach goes away obviously has other

02:02:29.740 people they work with and kind of implement things. But there's basically various different ways to

02:02:36.300 get the driver ready to react and have an optimal reaction speed off the line. But then you also

02:02:45.100 want to balance that against how they might react later in the race, the arousal curve we talked about

02:02:50.460 earlier. So there were various things that we tried in terms of training reaction time, there were some

02:02:55.420 supplements we tried out. So things around caffeine timing and dose tyrosine is an interesting one that

02:03:01.580 that may also be beneficial. I think we tinker with that a little bit. Creatine was obviously

02:03:05.180 something that came into play. There are other skills that I think translate across that kind

02:03:09.900 of thing like playing the drums. So you have to be relaxed, but you also have to be you know,

02:03:13.660 timing is critical. I don't think that one was ever implemented. But that was a thought process

02:03:18.380 that we talked about. There are a whole bunch of things that happened and a whole bunch of things

02:03:22.540 that were changed and the outcome improved. I can't say that anything that I did made the difference,

02:03:27.260 but at least we reached the end goal, which was a nice process of sort of scientifically tinkering

02:03:32.140 with things to get our end goal. Well, Tommy, thank you very much for humoring my endless

02:03:39.580 fascination with F1. And thank you more than that for making time to share with me and everybody

02:03:44.700 listening all of these insights. This is a fascinating discussion. I seem that the more I learn about the

02:03:51.500 brain, the less I know, which is a common refrain for anybody, I guess, who's trying to get deeper and

02:03:56.300 deeper into subject matter. So anyway, thank you very much, Tommy. And I hope to see you this year

02:04:01.500 at CODA. Yeah, likewise. Thanks so much for having me. I certainly feel the same about the brain. And

02:04:08.140 that's one of the reasons why I'm so fascinated with it. And I really appreciate the time to talk

02:04:13.180 to you. And then obviously, everybody for listening as well. Thanks, Tommy. Thank you for listening to this

02:04:17.740 week's episode of The Drive. If you're interested in diving deeper into any topics we discuss, we've

02:04:22.540 created a membership program that allows us to bring you more in-depth, exclusive content without

02:04:27.420 relying on paid ads. It's our goal to ensure members get back much more than the price of the

02:04:32.540 subscription. Now to that end, membership benefits include a bunch of things. One, totally kick-ass,

02:04:38.060 comprehensive podcast show notes that detail every topic, paper, person, thing we discuss on each

02:04:43.580 episode. The word on the street is nobody's show notes rival these monthly AMA episodes or ask me

02:04:50.060 anything episodes, hearing these episodes completely access to our private podcast feed that allows you

02:04:55.580 to hear everything without having to listen to spiels like this, the qualities, which are a super short

02:05:01.580 podcast that we release every Tuesday through Friday, highlighting the best questions, topics,

02:05:06.220 and tactics discussed on previous episodes of The Drive. This is a great way to catch up on previous

02:05:11.660 episodes without having to go back and necessarily listen to everyone. Steep discounts on products

02:05:17.420 that I believe in, but for which I'm not getting paid to endorse, and a whole bunch of other benefits

02:05:22.140 that we continue to trickle in as time goes on. If you want to learn more and access these member-only

02:05:26.940 benefits, you can head over to peteratiamd.com forward slash subscribe. You can find me on Twitter,

02:05:33.580 Instagram, Facebook, all with the ID peteratiamd. You can also leave us a review on Apple Podcasts or

02:05:40.620 whatever podcast player you listen on. This podcast is for general informational purposes only and does

02:05:46.620 not constitute the practice of medicine, nursing, or other professional healthcare services,

02:05:51.420 including the giving of medical advice. No doctor-patient relationship is formed. The use

02:05:56.940 of this information and the materials linked to this podcast is at the user's own risk. The content on

02:06:03.020 this podcast is not intended to be a substitute for professional medical advice, diagnosis, or treatment.

02:06:09.740 Users should not disregard or delay in obtaining medical advice from any medical condition they have,

02:06:16.380 and they should seek the assistance of their healthcare professionals for any such conditions.

02:06:21.820 Finally, I take conflicts of interest very seriously. For all of my disclosures and the companies I invest in

02:06:27.580 or advise, please visit peteratiamd.com forward slash about where I keep an up-to-date and active list of such companies.

02:06:39.020 See you all the time and forward slash about where I keep an up-to-date and active list of you.

02:06:43.260 See you in your case.

02:06:45.980 Bye.

02:06:47.380 One thing that you've been talking about.

02:06:49.660 What about you?

02:06:51.180 That's finally fun.

02:06:53.280 Bye.

02:06:54.180 Bye.

02:06:54.740 Bye.

02:06:55.080 Bye.

02:06:55.240 Bye.

02:06:55.900 Bye.

02:06:57.200 Bye.

02:06:57.360 Bye.

02:06:57.900 Bye.

02:06:58.080 Bye.

02:06:58.720 Bye.

02:07:03.920 Bye.

02:07:06.240 Bye.

02:07:06.800 Bye.

02:07:07.940 Bye.

The Peter Attia Drive - June 05, 2023

#257 ‒ Cognitive decline, neurodegeneration, and head injuries： mitigation and prevention strategies, supplements, and more ｜ Tommy Wood, M.D., Ph.D.

Episode Stats

Summary

Transcript