The Peter Attia Drive - #41 - Jake Kushner, M.D.： How to thrive with type 1 diabetes and how everyone can benefit from the valuable insights

00:00:00.000 Hey everyone, welcome to the Peter Atiyah drive. I'm your host, Peter Atiyah. The drive

00:00:10.880 is a result of my hunger for optimizing performance, health, longevity, critical thinking, along

00:00:15.940 with a few other obsessions along the way. I've spent the last several years working

00:00:19.660 with some of the most successful top performing individuals in the world. And this podcast

00:00:23.620 is my attempt to synthesize what I've learned along the way to help you live a higher quality,

00:00:28.360 more fulfilling life. If you enjoy this podcast, you can find more information on today's episode

00:00:33.000 and other topics at peteratiyahmd.com. Hey everybody, welcome to this week's episode

00:00:43.360 of the drive. I'd like to take a couple of minutes to talk about why we don't run ads on this podcast

00:00:48.580 and why instead we've chosen to rely entirely on listener support. If you're listening to this,

00:00:53.820 you probably already know, but the two things I care most about professionally are how to live

00:00:59.060 longer and how to live better. I have a complete fascination and obsession with this topic. I

00:01:04.760 practice it professionally and I've seen firsthand how access to information is basically all people

00:01:09.920 need to make better decisions and improve the quality of their lives. Curating and sharing this

00:01:15.100 knowledge is not easy. And even before starting the podcast, that became clear to me. The sheer volume

00:01:20.380 of material published in this space is overwhelming. I'm fortunate to have a great team that helps me

00:01:25.800 continue learning and sharing this information with you. To take one example, our show notes are in a

00:01:31.600 league of their own. In fact, we now have a full-time person that is dedicated to producing those

00:01:36.260 and the feedback has mirrored this. So all of this raises a natural question. How will we continue

00:01:42.440 to fund the work necessary to support this? As you probably know, the tried and true way to do this

00:01:47.980 is to sell ads. But after a lot of contemplation, that model just doesn't feel right to me for a few

00:01:54.200 reasons. Now, the first and most important of these is trust. I'm not sure how you could trust me if I'm

00:02:00.400 telling you about something when you know I'm being paid by the company that makes it to tell you about

00:02:05.080 it. Another reason selling ads doesn't feel right to me is because I just know myself. I have a really

00:02:11.280 hard time advocating for something that I'm not absolutely nuts for. So if I don't feel that way

00:02:16.460 about something, I don't know how I can talk about it enthusiastically. So instead of selling ads,

00:02:21.580 I've chosen to do what a handful of others have proved can work over time. And that is to create

00:02:27.420 a subscriber support model for my audience. This keeps my relationship with you both simple and

00:02:33.480 honest. If you value what I'm doing, you can become a member and support us at whatever level

00:02:39.460 works for you. In exchange, you'll get the benefits above and beyond what's available for free.

00:02:44.300 It's that simple. It's my goal to ensure that no matter what level you choose to support us at,

00:02:50.040 you will get back more than you give. So for example, members will receive full access to the

00:02:57.520 exclusive show notes, including other things that we plan to build upon, such as the downloadable

00:03:04.240 transcripts for each episode. These are useful beyond just the podcast, especially given the technical

00:03:09.180 nature of many of our shows. Members also get exclusive access to listen to and participate

00:03:15.940 in the regular ask me anything episodes. That means asking questions directly into the AMA portal

00:03:22.660 and also getting to hear these podcasts when they come out. Lastly, and this is something I'm really

00:03:28.000 excited about. I want my supporters to get the best deals possible on the products that I love.

00:03:32.820 And as I said, we're not taking ad dollars from anyone, but instead what I'd like to do is work

00:03:37.360 with companies who make the products that I already love and would already talk about for free and have

00:03:43.220 them pass savings on to you. Again, the podcast will remain free to all, but my hope is that many of

00:03:51.000 you will find enough value in one, the podcast itself, and two, the additional content exclusive

00:03:57.900 for members to support us at a level that makes sense for you. I want to thank you for taking a moment

00:04:02.960 to listen to this. If you learn from and find value in the content I produce, please consider

00:04:08.480 supporting us directly by signing up for a monthly subscription. My guest this week is Dr. Jake

00:04:13.880 Kushner, a close friend and a pediatric endocrinologist who specializes in helping

00:04:18.760 manage people with type one diabetes. In this episode, we go into great detail about type one

00:04:24.400 diabetes. Probably the biggest takeaway from this episode is that diet can help people not only with

00:04:29.920 type two diabetes, but also type one. Jake discusses some of the recent findings in an approach

00:04:35.220 that one of the pioneers in this field, Dr. Richard Bernstein championed for several decades now

00:04:40.000 managing carbohydrates and protein to manage insulin and glucose. And if you're thinking, well,

00:04:45.380 I don't have type one diabetes. Why does this matter to me? The answer is anybody with a pancreas

00:04:50.620 is going to benefit from this. And obviously those without a pancreas, which is effectively what is

00:04:55.560 happening in type one diabetes. We also discuss why many patients who control their blood sugar

00:05:00.180 with high amounts of exogenous insulin are at substantially higher risk than people who can

00:05:04.820 control their blood sugar at lower levels of insulin. This concept also has implications for

00:05:09.620 everyone else, as I said, because the word exogenous and endogenous can be interchanged quite

00:05:14.360 easily when it comes to insulin. We cover some of the basics, the history, the increasing prevalence of

00:05:19.660 type one diabetes, and more importantly, what we can do to help kids with this disease and their

00:05:24.000 families and the non-diabetics out there learn from them and their disease. So without further 1.00

00:05:29.700 delay, here's my conversation with Jake Kushner. Hey Jake, thanks for coming over, man.

00:05:37.220 Hey Peter, thanks for having me. It's a pleasure.

00:05:39.980 What do you think is the probability we will get through this discussion today without one of my kids

00:05:43.540 barreling into this room? Absolutely zero, especially given that I met both of them and they're

00:05:48.660 incredibly rambunctious and cute. Well, you met two boys, but there's a girl too.

00:05:52.340 Okay. Oops. She's much more well-behaved. So you're in town. I guess we can't talk about it.

00:05:58.400 It's Black Ops Mission, right? No, it's actually, so I'm an advisory council for Sanofi and I've been

00:06:04.660 an advisor about this drug, sodagliflozin, which is part of a new class of drugs that will hopefully

00:06:10.680 benefit people with type one diabetes, these adjuvant therapies, molecules that you take in addition to

00:06:16.700 insulin. And I've been involved in this program now for close to seven years.

00:06:20.380 And I know off mic, we've spoken a little bit about some of those compounds. They're super

00:06:24.680 interesting, not just for type one diabetes, but potentially type two diabetes. And frankly,

00:06:28.880 if you subscribe to the idea that lower levels of glucose and insulin are just better overall,

00:06:32.440 there may even be other applications. So before we get into that, we met, it's been three or four

00:06:38.820 years now, right? We met in Houston over dinner one night and I gave you credit for this recently on

00:06:45.480 a podcast that hasn't yet released, but it will release. It was actually you who first told me

00:06:50.700 about CGM, although it was officially Kevin Sayre at Dexcom that got me fully hooked. So whenever I'm

00:06:59.320 talking to somebody and they ask me about a question that has to do with type one diabetes,

00:07:03.680 I generally say like, I know very little about this condition, but you should meet Jay Kushner because

00:07:09.120 not only does he get it, but I think his outcomes are pretty unique. I think we're seeing more people

00:07:15.480 move towards the realization that you and people who think like you have, but so you're a pediatric

00:07:20.120 endocrinologist. Did you always know you wanted to do that?

00:07:23.240 Well, I was trying to decide in between being a pediatrician and being a molecular biologist.

00:07:28.020 And my parents are basic scientists and, you know, there was a well-worn path in between

00:07:32.540 molecular biology and endocrinology. So that was a really typical thing back in the mid eighties.

00:07:38.220 And I thought I would do that. And the classical path as an endocrinologist is to go and find a

00:07:44.320 rare disorder and discover the molecular basis for it and find some unique product that you could then

00:07:49.820 replace and provide it. And then you're good to go. And there's a lot of endocrinologists who

00:07:54.880 made quite a bit of hay on that paradigm. And so I was also interested in pediatrics. I was trying to

00:08:02.100 decide between all these things when I was an undergraduate at Berkeley and working pretty much

00:08:06.280 continuously in basic science research labs all the way through Berkeley. And then for several

00:08:11.760 years afterwards as a research technician, first at UCSF and then Oregon. So I applied to medical

00:08:17.200 school and I get in and I think I'm going to be a pediatrician. And my mother was greatly disappointed

00:08:22.840 that I didn't become a basic scientist. She viewed science as much more compelling than becoming a

00:08:28.780 pediatrician.

00:08:29.380 Yeah. Or any doctor for that matter, right? I mean, there could be no greater letdown to,

00:08:34.860 you know, scientific parents that you would choose something as low as medicine.

00:08:39.600 And I was literally the only kid in medical school whose mother was disappointed that I didn't

00:08:44.320 end up going to graduate school.

00:08:46.460 If it makes you feel any better, my dad was very disappointed that I went to medical school too.

00:08:50.540 So I found myself there. I thought I was going to be a pediatrician. I did my clinical

00:08:55.260 rotations and that it totally reinforced it. And I just love being on the wards and interacting with

00:09:00.700 kids. And then ended up as a pediatric resident at Brown and had, again, I'd worked in molecular

00:09:06.840 endocrinology as a researcher before I went to medical school and was enamored by endocrinology in

00:09:13.840 general. And I applied and got into this program at Boston Children's. So there I was in this incredibly

00:09:19.480 busy, rigorous, unbelievably complicated place. And that's when this, this crazy thing happened in

00:09:27.900 my life. And so I thought I was going to find some esoteric metabolic disease to work on.

00:09:32.160 And you were at the right place to do it because for the folks listening who might not know this,

00:09:36.480 there are probably what, four major pediatric centers in North America, Boston Children's,

00:09:42.000 obviously in Boston, Sick Kids in Toronto, CHOP in Philly, and maybe Texas Children's. Would that be in

00:09:48.180 the top four? Yeah, in UCSF. That's right. Would UCSF actually make the top five? At the time,

00:09:54.100 it did. Okay. Incredibly strong. So I sort of thought I would do this and I'd find some esoteric

00:09:59.860 disease, like maybe work on calcium or, you know. But as a first-year endocrine fellow, we're running

00:10:05.640 around an incredibly busy service and we're interacting with families of children who are

00:10:11.620 newly diagnosed. And it's such a compelling- With type 1 diabetes.

00:10:15.380 Type 1 diabetes. And it's so compelling. So we would take call, the ER would call us,

00:10:21.060 oh, there's a child who's been diagnosed. We come in the next morning and meet these families.

00:10:25.800 And, you know, there's a tearful, exhausted, overwhelmed mother. There's a kid who literally

00:10:30.940 looks like she's been like shot out of a cannon, you know, with multiple IVs and scars in her arms from 1.00

00:10:37.920 blown IVs. And we have to try to work through this idea that they were going to be insulin

00:10:43.800 dependent for life. And I just thought, wow, this is really an amazing and challenging problem.

00:10:50.300 And it has so many aspects of medicine that I appreciate as most importantly, human elements,

00:10:56.300 but there's also basic science.

00:10:57.780 The diagnosis is, I mean, I had one experience with this before I got to medical school,

00:11:02.460 which was in college. I had a really good friend. I guess I won't use her name just in case I need

00:11:07.400 to protect her identity. And she was amazing. She was my lab partner. We're just great, great

00:11:13.520 buddies. She was a stud athlete. She was in the rowing team. She was like the perfect physical 0.79

00:11:18.840 specimen of, you know, and we used to sit together and we were in two classes together. And she was

00:11:25.080 like maybe four or five months into that school year and she couldn't stay awake anymore. She was like

00:11:29.960 falling asleep in class constantly. And I remember sort of nudging her and being like, Hey,

00:11:35.660 you're burning it too much. Like the practices are killing you and the labs and blah, blah, blah,

00:11:40.340 blah. Like you got to get this shit under control. And she's like, no, no, no, I'm fine. I'm fine. 1.00

00:11:44.800 I'm fine. She just couldn't stay awake to save her life. And she was losing weight like crazy.

00:11:49.280 And she was lean to begin with. So she was, you know, probably six feet tall and might've weighed 0.99

00:11:55.340 140 was her fighting weight. And, you know, she was now down to 125 or something like that. And of

00:12:01.640 course she was going to the bathroom all the time. And of course I'm just a dumb engineer. I don't 0.99

00:12:06.240 know anything. I don't realize this is such an obvious, you know, sign of somebody who's going

00:12:11.540 to be diagnosed in a medical school and you can't die. But it's really difficult for everybody.

00:12:15.380 Oh, this was in an, it's an undergrad. Oh yeah. This is before I even knew I wanted to go into

00:12:18.880 medicine. But yeah, sure enough, she comes back one day with, and she's like, Oh my God,

00:12:23.000 you're not going to believe this. I finally went to the doctor and all of a sudden for

00:12:26.400 the rest of school, she takes insulin. And for the rest of her life. Yeah. And these

00:12:30.800 are previously healthy people. They're cruising around, living their lives. There's nothing

00:12:34.640 obvious that contributes to the risk. And then they start drinking a lot and peeing a lot and

00:12:40.300 they lose weight and they essentially are in a state of profound catabolism and their bodies

00:12:47.520 waste away because insulin is such a really pivotal signal for directionality on metabolism.

00:12:53.000 And they decide to break it down. And it's amazing. It beguiles everyone who's up there

00:12:59.120 close. There are so many misdiagnoses around type one diabetes. About 30% of children still

00:13:05.220 present with DKA. That is to say they've had prolonged illness and no one's thought that

00:13:10.140 they could actually be quite ill.

00:13:11.680 So when, to be clear, when they present with diabetic ketoacidosis, it means that they have

00:13:17.060 somehow managed to slip through the cracks. Nobody's figured out that their glucose level is

00:13:22.140 through the roof or that they're becoming acidotic or any of these other things. And what actually

00:13:26.420 brings them into the hospital is a septic-like situation where their, you know, their blood

00:13:30.780 pressure is falling and they're becoming unresponsive.

00:13:33.360 Yeah. They hit the metabolic wall. And at that point, they're only on this earth for a matter

00:13:39.060 of days or a week or two without insulin. And it's quite common for very young children to present

00:13:46.060 in DKA. It's less common in older people. They typically have more compensatory reserves. 0.57

00:13:52.040 So they have more, the autoimmunity proceeds at a slower rate as you get older. And so the diagnosis

00:13:58.560 becomes more and more confusing the older the onset of the disease. And this phenomenon actually

00:14:04.600 progresses into adulthood to the point where there are people our age who are diagnosed with diabetes,

00:14:10.000 diabetes, not otherwise specified, who think they have type 2 and they're skinny, but in fact have

00:14:16.720 cryptic type 1 diabetes. They probably have autoimmune antibodies and they have some degree of a loss of

00:14:23.220 beta cells. And this phenomenon explains some of the adult onset diabetes that you just don't quite

00:14:29.360 understand. How is this person diabetic?

00:14:31.840 And so fast forward a few years from where I am, and I want to come back to where you were, but by the

00:14:36.480 time I'm in medical school, I had only been exposed to people who had type 1 diabetes, which is ironic

00:14:42.900 given the prevalence difference between type 1 and type 2. And I remember in a pathology class in my

00:14:47.800 first year, they said, well, you know, I forget what the number was at the time, but they said about 95%

00:14:53.300 of the cases of diabetes are what we call type 2 or adult onset diabetes. And I remember thinking to

00:14:59.980 myself, the professor must've made a mistake. I've never seen this type 2 diabetes thing they talk about,

00:15:05.000 but I've seen a lot of people, not just that friend in school, but I went on to see other people and

00:15:09.580 knew of other people, though I didn't watch them through the diagnosis. So what is the prevalence

00:15:13.460 today, notwithstanding this late onset type 1 diabetes that we'll come and talk about later,

00:15:19.340 but just directionally, how many people in the United States have type 1 versus type 2 or

00:15:24.540 indeterminate?

00:15:25.600 Okay. So type 1 is approximately 1 in 300 people. And the typical age of onset is around 9. And

00:15:31.100 they live into adulthood and, and many survive into their eighties or even nineties. They have,

00:15:38.340 of course, reduced survival because of complications, but it's an incredibly common

00:15:42.540 illness. It's the most common life-threatening medication requiring illness of childhood.

00:15:48.620 Yeah. Which is amazing when you go back to where you were at Boston Children's Hospital,

00:15:51.780 I'm sure for any pediatric endocrinologist to be working on glycogen storage diseases or

00:15:56.720 inborn areas of metabolism is interesting intellectually, but the scale upon which those

00:16:02.980 diseases afflict children is trivial compared to what you just described.

00:16:07.520 And when you're working in a fancy medical school, it's quite sort of tempting to focus on a rare,

00:16:13.360 unique condition that, that most people don't get a lot of exposure to and sort of bypass the

00:16:19.420 really common things. But for me, type 1 diabetes was the elephant in the room. It was just so

00:16:24.540 compelling. And I ended up being the first clinical fellow at Boston Children's in 10 years to go into

00:16:31.240 type 1 diabetes.

00:16:32.460 Were any of your mentors there disappointed in the way that your mom was that you were sort of,

00:16:36.500 I don't want to say slumming it because I don't mean to suggest that that's what it was,

00:16:40.080 but that you were stepping off the pedestal of the esteemed academic institution?

00:16:44.720 No, I ended up working on type 1 diabetes at the Jocelyn Diabetes Center, which was an amazing place.

00:16:50.220 Also in Boston.

00:16:51.960 Also in Boston. There was a long August tradition of academics there. They were a little surprised

00:16:57.440 that I was interested in going after such a tough problem. And again, clinically, type 1 diabetes is

00:17:02.380 an immense problem, as we'll get into, because the clinical treatments are just so difficult and

00:17:08.080 ineffective and variable in their response.

00:17:10.280 So let's talk a little bit about the pathophysiology of this. Presumably people listening to this who already

00:17:14.740 know everything about it and want to get to the fun stuff, we'll skip this. We'll timestamp it so they can

00:17:19.180 jump ahead. But I think it is worth understanding because there's definitely some confusion.

00:17:23.300 You obviously alluded to the fact that this has an autoimmune component, if not the main component.

00:17:27.960 So what, based on everything we know today, is the pathophysiology of this condition?

00:17:33.960 Okay. Well, I guess we could say just from a very high level, there are clearly genetic as well as

00:17:39.520 environmental contributors. And so we know that it runs in families, and we know that certain regions or

00:17:45.820 certain behaviors predispose. But if you actually get down to the molecular pathophysiology,

00:17:51.160 what you see is that it's an autoimmune condition, largely driven by T cells. But there is also a

00:17:57.520 contribution of B cells. And because of that, you can detect antibodies in the serum that indicate

00:18:03.940 autoimmunity. And so we generally believe what happens is, by some combination of bad luck,

00:18:11.880 and possibly genetics, and possibly the foods you eat to a very, very modest degree, this stochastic

00:18:19.260 phenomenon begins to move forward, whereby the beta cell, which secretes insulin, begins to dump

00:18:25.400 off antigens. And the antigens begin to provoke the immune system locally. And some of those immune

00:18:31.820 cells start to destroy beta cells and dump out more antigens.

00:18:35.020 So let me take a step back for the listener. The pancreas, of course, is this gland. 0.50

00:18:39.180 Yeah, thank you.

00:18:39.640 No, no, no, that's fine. The pancreas is this gland that sits in the retroperitoneum, 0.50

00:18:43.520 so it's not in the abdomen proper, but it's behind the stomach. And about 95% of the pancreas by mass

00:18:50.680 serves this exocrine function, which is it's mostly there for local digestion. It puts enzymes into the

00:18:58.740 biliary stream that ultimately aid in the proximate digestion if things come out of the stomach. 0.54

00:19:02.880 However, about 5% of the pancreas by mass, including these cells called beta cells,

00:19:07.300 but presumably you'll tell us a little bit about alpha and delta cells. It's an endocrine organ

00:19:11.840 versus an exocrine organ, which means it's secreting systemic hormones into the system.

00:19:17.520 So I remember learning that in medical school and being surprised because I'd heard so much

00:19:21.960 about beta cells. And I was like, wait, they only make up 5% of this pancreas? Like,

00:19:26.220 they're a relatively small contribution to this enormous organ, right?

00:19:29.660 And essentially insulin is only made there, except perhaps in tiny quantities in the brain.

00:19:35.080 And so the, yes, these islands of cells, these islets of Langerhans exist floating in a sea of

00:19:40.800 the acinar or exocrine component, and they make insulin uniquely, unlike any other tissue.

00:19:47.360 And perhaps because they make these unique antigens or these unique cellular products like insulin and

00:19:52.880 some of the other apparatus that are required for glucose-stimulated insulin secretion,

00:19:59.100 when they begin to dump off their antigens and provoke autoimmunity, you can get this sort of

00:20:05.520 death spiral where ultimately the end result is a progressive autoimmunity where you lose beta cells

00:20:12.960 over time. And I don't think that there's any one single, for instance, virus that can be

00:20:20.100 conclusively ascribed to autoimmunity and type 1 diabetes. It's clearly not related to,

00:20:27.220 directly related to your BMI or your body mass index or your body weight, though it's certainly

00:20:32.420 a contributor. And also genetics are-

00:20:35.060 Which way does it seem to contribute, high or low?

00:20:37.440 In populations, again, it's not, it doesn't hold up in individuals very well, but in populations,

00:20:43.440 the heavier you are, the more likely you are to have type 1, which implies either that the work of the

00:20:49.600 beta cells contributing to the autoimmunity, or alternatively, the work of the beta cell and the

00:20:55.700 relative beta cell insufficiency reveals this autoimmunity that otherwise would have been cryptic.

00:21:02.000 We've seen a rise in the incidence, and of course, then the prevalence of type 2 diabetes.

00:21:07.180 That's very difficult to ignore, and I've written about that in other places, and you can chase that

00:21:12.380 all the way back to the late 1800s, and it tracks pretty well. What has been the relative change in

00:21:18.680 the incidence of type 1 diabetes?

00:21:20.980 So it's doubled since 1960, and it appears to be continuing to increase. And so one possibility is

00:21:28.080 that there are these environmental factors that are increasing, for instance, obesity. Alternatively,

00:21:32.740 there could be viruses or other things that we're exposed to. And another possibility is that we're

00:21:37.540 simply better at diagnosing it. So we know for a fact that some infants that are diagnosed now with

00:21:42.860 new onset diabetes would have been the kids who you thought had died of sepsis back when you were in

00:21:48.380 medical school. So we have a greater awareness that this autoimmunity can happen to kids as early

00:21:55.120 as five or six months of life. And at that point, they don't present with the classical signs and

00:22:01.200 symptoms. It's really incredibly aggressive. They look like they have gastroenteritis and weight

00:22:07.140 loss. And if nobody bothers to check a serum glucose, you might never know that they in fact

00:22:13.540 have life-threatening diabetic ketoacidosis and new onset type 1 diabetes. It's a disaster.

00:22:18.480 I suspect it was because I knew we were going to be speaking today. But last night when I was putting

00:22:22.660 my littlest guy to bed, who's about 16, 17 months old, I remember I was looking at him and I was

00:22:29.880 playing with him and I'm biased. So I think he's as cute as all hell. And I remember thinking,

00:22:34.260 what would happen if this guy got type 1 diabetes? How would we figure it out? How long would it take

00:22:41.480 us to figure out, oh my God, check his glucose? And it scared me actually, because it's like,

00:22:48.860 I think about this stuff night and day and I realized it would easily escape me for a long time.

00:22:55.060 And to your point, they don't have the reserve. They don't have the physiologic reserve,

00:22:59.600 which is just doctors speak for. They don't have the buffer between life and death that you or I

00:23:05.440 would have. And they actually might have much more aggressive autoimmunity. 0.66

00:23:09.440 Meaning it comes on quicker. The rate of decline of endogenous insulin production is much faster.

00:23:16.180 Yeah. They might present with 90% of their beta cells gone because the autoimmune destruction from

00:23:22.940 100% of their beta cell capacity to 10% could occur in a matter of months. Almost by definition,

00:23:30.100 must in a kid who's six months old. But if I'm 53, if I got type 1 diabetes, I could smolder along

00:23:36.780 for years. And again, I have a friend who's my age, who is essentially my height and weight,

00:23:44.920 and he has new onset diabetes. And we think he actually has cryptic type 1. And it's slowly

00:23:51.680 smoldering. It seems to be slowly progressing. It's been very difficult to precisely make the

00:23:57.820 diagnosis.

00:23:58.340 And I want to come back to that in time for at least one reason, which is for people out there

00:24:04.120 listening, there are people getting caught between these crosshairs that are potentially being

00:24:09.280 misdiagnosed. Outside of the intellectual exercise of getting it right, which we love to do as doctors,

00:24:16.320 are there treatment implications that are more relevant? In other words, if you have someone

00:24:21.260 who's walking around with quote unquote type 2 diabetes, but they're really late onset type 1,

00:24:26.700 are we also potentially caring for them incorrectly? So I'll park that because there's so much I want

00:24:31.980 to go back to in type 1. But between the two of us, let's try to remember to come back to that.

00:24:36.520 So I didn't actually realize the B cells played a big role, but I guess that makes sense. So I thought 1.00

00:24:42.140 it was just a CD4, CD8, CD25 problem, but it's beyond that.

00:24:47.080 Yeah, there's this idea of an antigenic spread. So again, if the beta cells make very unique things,

00:24:53.080 and when you break them open and destroy them, you begin to dump them out, there is both a B

00:24:57.960 and a T cell response. And we know this because we can test for antibodies in the serum of people

00:25:03.980 type 1 diabetes. And they make unique antibodies against insulin and GAD, glutamic acid decarboxylase,

00:25:11.920 as well as a protein tyrosine phosphatase called ICA512. And so those antibodies are commonly used

00:25:19.820 in the diagnosis of type 1 diabetes, which we believe to be a T cell disease. But again,

00:25:24.340 that doesn't make any sense because antibodies are made by B cells. So there must be some contribution.

00:25:28.540 And the definitive proof of that was in a New England Journal article that was done by my friend,

00:25:35.860 the late Mark Peskovitz. And he used, in a huge clinical trial, rituximab, which alters B cell

00:25:42.460 function, alters the ability of the B cells to make antibodies. And they were able to delay the course

00:25:49.540 of type 1 diabetes.

00:25:50.840 So they were able to intervene in a subset of patients that were progressing slowly enough.

00:25:55.760 They presented with positive GAD antibodies, but they still had beta cell production.

00:26:00.240 You whack them with rituximab. You at least take the B cell component out of this.

00:26:04.600 And if the control patient isn't having that, you now create a Kaplan-Meier curve for a beta cell.

00:26:09.400 That's right. That's right. And so they were only temporarily able to alter.

00:26:13.440 How much did they delay the onset?

00:26:15.360 So rituximab is quite potent. So the circulating antibodies were reduced dramatically.

00:26:20.000 They only gave a single set of infusions. And if I understand, remember this study correctly,

00:26:25.040 they got a response, but it was not statistically significant at a year.

00:26:29.760 Based on one single dose?

00:26:31.360 I'm not sure about that.

00:26:32.300 We'll pull it up.

00:26:33.000 Yeah.

00:26:33.460 Okay. Are there any other autoimmune diseases that share this phenomenon that you're aware of,

00:26:38.200 where you have this B cell, T cell component?

00:26:40.900 Well, I think it's true of a bunch of the other autoimmune endocrine disorders,

00:26:44.620 such as, for instance, autoimmune thyroid disease, where you can get thyroid peroxidase and

00:26:49.120 thyroglobulin antibodies. And again, we largely think that that's a T cell mediated disease. And

00:26:55.140 by the way, autoimmune thyroid disease is associated with type 1 diabetes or vice versa.

00:27:00.580 Meaning patients who will go on earlier in life to develop type 1 diabetes would be

00:27:04.800 much more likely than patients without it to go on to get Graves disease or some of the other

00:27:08.840 autoimmune thyroid disorders?

00:27:10.440 Yeah. Typically type 1 would present first and then they can get autoimmune thyroid disease,

00:27:14.180 but they can also get adrenal disease. And we also see, for instance, rarely sciulitis,

00:27:19.720 which is autoimmune disease of the salivary gland, and even rarely autoimmune euphritis of the ovaries,

00:27:26.860 as well as pernicious anemia, the ability to produce B12. And so that's presumably an

00:27:32.900 autoimmune disease of the cells that make B12.

00:27:35.980 Wow. I didn't actually realize that there was so much overlap between these. Let's clarify one

00:27:40.460 other thing that I'm sure someone's wondering. One thing that parents, I think,

00:27:43.660 are pretty mindful of is what do we have to feed our kids to minimize the risk of them getting

00:27:48.680 anaphylactic reactions to nuts or things like that, given the consequences of such things.

00:27:53.500 Not to get off into the weeds, but it's probably worth explaining why that's a completely different

00:27:58.100 process than what we're talking about with the autoimmunity with respect to the endocrine. So

00:28:02.580 not to put you on the spot, and I'm happy to fill in the gaps, but what is it about a kid when they

00:28:07.440 eat a peanut that leads to an anaphylactic reaction? And why is that a totally different type of immune

00:28:11.620 response? Well, I should just say, in terms of anaphylactic reactions for peanuts, it turns out

00:28:16.180 that you probably want to eat peanuts to avoid the anaphylactic reaction. Isn't that amazing?

00:28:20.400 That is certainly what we believe now, right? I think the literature has become crystal clear

00:28:24.400 that early exposure is better. Right.

00:28:26.800 But again, that's a very different type of immune response, isn't it? Yes.

00:28:29.560 That's mediated by IgE. Right.

00:28:31.800 That's a hypersensitivity more than an autoimmune response. Is that correct?

00:28:35.440 Yeah. So borrowing from that though, do we believe that there are any exposures that are

00:28:42.420 withheld from kids that may be predisposing them to beta cell destruction? Or do we think it has

00:28:48.340 much more to do with these diffuse genetic issues, which I want to come back to these quote unquote

00:28:53.000 environmental issues. And then of course, what it seems to be the majority of it seems to be

00:28:56.720 stochastic. So there's no one precise thing. And moreover, I don't believe that parents who are

00:29:02.440 listening to this could say, okay, I want to make sure that my kid doesn't get type 1 diabetes. So

00:29:06.200 we're not going to eat any enriched carbohydrate, and then we'll mitigate that risk completely.

00:29:11.760 However, that said, in populations, you can see these very subtle differences in type 1 diabetes

00:29:19.460 risk, depending upon the body weight of children, which implies that the standard American diet may

00:29:25.180 be predisposing the population in aggregate ever so slightly towards more type 1 diabetes,

00:29:31.160 which is a scary thought.

00:29:33.220 Yeah. The challenge, of course, is because the environmental contribution is so much smaller

00:29:37.620 here than it is in type 2 diabetes, we don't know if we could be looking at genetic differences that

00:29:42.540 could explain that. In other words, we can't do the experiment as easily the epidemiologic

00:29:47.180 comparison, I guess, to be more rigorous in my terminology, of comparing a US population to,

00:29:53.000 you know, pick a completely different population that is genetically completely distinct

00:29:57.580 and environmentally distinct. We could get fooled by that, right?

00:30:00.960 Yeah, but there is a pretty nice study using the T1D exchange data, which is a large consortia of

00:30:08.080 clinics around the United States. And in fact, there's a paper that was published by Maria Redondo

00:30:13.780 and several other friends of mine describing explicitly these relationships in between body weight

00:30:19.260 and the potential risk for type 1. So that's interesting.

00:30:22.080 And do you remember just directionally when a child goes, I don't even know kid body BMI is like

00:30:27.240 my little guy who you just saw, who's like 97th percentile in height, weight, and 95th percentile

00:30:32.680 in head circumference. I was looking at his little pediatric sheet the other day because he went to the

00:30:37.320 doctor and he was like BMI of 19.4. So kid BMIs mean, I don't know how to think in those BMIs,

00:30:43.460 but I was going to say like, what's the difference between a BMI 16 and a BMI 20? Like how much of

00:30:49.640 an increase in relative risk do you see? Pediatricians like me never are able to memorize

00:30:55.140 these things. You've got to put it on a, you know, we have to plot it on a chart. So there's growth

00:30:59.640 curves for both height and weight and head circumference and weight versus height or also

00:31:05.740 body mass index. It's just so complicated and there's different curves for boys and girls.

00:31:10.240 Yeah. Okay. So question one out of the way is, is there anything I can do or a parent can do to

00:31:18.000 reduce the risk of our kids getting type one diabetes? The answer is probably not other than

00:31:23.660 if you want the wind a little bit at your back, as opposed to a little bit in your face, if your kids

00:31:28.160 can be of closer to normal BMI than not. Being aware that it is a possibility and being able to

00:31:34.120 inform yourself and your neighbors and in your school teacher undiagnosed type one diabetes is

00:31:41.100 very dangerous for children. And we need to do a better job of detecting it early.

00:31:45.040 Yeah. With one in 300 kids getting it, that is, I mean, that's, that means in any given school,

00:31:51.120 one to three kids are being diagnosed a year, which is interesting because I just don't have that

00:31:55.020 recollection. You know, I, but it may speak to this issue you've talked about, which is we just

00:31:59.240 underdiagnosed. We've sort of talked briefly about genes, but obviously this is not going to fall in

00:32:05.060 the category of single gene mutations like the glycogen storage diseases or some of the other

00:32:09.560 errors of inborn metabolism. Do we have even a quasi sense of how heavy the genetic contribution is

00:32:17.280 and what those genes may actually do from a functional standpoint?

00:32:21.620 We understand it pretty well. And though there's, there's still a lot to be learned. So it's complicated

00:32:27.260 in that there's essentially two different themes of the genetics of type one diabetes. One are these HLA

00:32:34.260 alleles, like this is the human histocompatibility locus. And there's, there's certain alleles that are

00:32:40.920 quite common in the population that when combined place people at higher risk. And so there's something

00:32:48.300 called HLA DR3 and DR4. And these alleles tend to run quite frequently in the U S population.

00:32:56.520 And as a result, people who have type one diabetes typically will have about a tenfold risk of having

00:33:03.780 a child with type one diabetes relative to the, to the general population.

00:33:07.020 And to be clear, is that assuming they inherit the same HLA type or independent of the inherited

00:33:13.200 pattern of their HLA? Cause if you're, if you are HLA to DR, your kid has a, is that, is that a

00:33:20.080 co-dominant inheritance? Is your kid 50% likely to get that? Assuming your wife doesn't have that? 0.96

00:33:24.380 Yeah. But what happens is it's so common that your wife may have that. And so just statistically,

00:33:29.900 because children end up having it, the people who have type one diabetes often have these alleles and

00:33:35.300 also these other minor alleles that contribute each very small amounts, but in aggregate quite a bit.

00:33:42.580 So the analogy I use is it's sort of like, let's play a poker game, if you will. We're going to play

00:33:48.240 a poker game where on your left hand, we're going to play 21. And on the right hand, we're going to

00:33:53.520 play five card stud. Okay. And so when you deal me, and by the way, the winner is the one.

00:34:00.280 So when you say 21, you mean blackjack on your left hand? 0.95

00:34:02.280 Exactly. Blackjack on my left hand and five card stud on the right. And when you're dealt the cards,

00:34:08.240 you have the HLA alleles, which is blackjack, because there's only a few and they sort out

00:34:12.680 however they sort out. And the right hand, you're playing five card stud. And in fact,

00:34:17.240 there's a huge number of combinations that could come in five card stud, but together those two

00:34:23.020 hands determine the risk. And so it's quite complicated because you can't draw a direct line

00:34:27.720 in between one gene or a group of genes. But in aggregate, the statisticians think that

00:34:32.760 that people are approximately six to tenfold at greater risk. If you have somebody who has type

00:34:39.360 one to have a direct relative. And so if you're someone with type one, that's still not a reason

00:34:43.720 to get completely panicked because that means you're, it's still unlikely your kid is going to

00:34:49.080 get type one diabetes, right? If you have a disease that's hitting one in 300 and you have a tenfold

00:34:53.740 increase, it's going to hit one in 30. There's still a 97% chance your child is not going to

00:34:59.920 have type one diabetes if you do, but it would have been a 99.7% chance, hopefully, if you hadn't,

00:35:05.780 if I'm doing my math right. But there are these families, I've met them, you know, I met one family

00:35:10.440 where all three sons had type one diabetes. Oh God. And was the age of onset comparable in each of these

00:35:18.540 children? It doesn't necessarily align. That's kind of an interesting and surprising element.

00:35:23.860 Yeah. And actually, I, at one point cared for identical twins who were discordant for type one

00:35:29.320 diabetes. Well, that's, I was going to ask you that. What is the twin concordance? Identical

00:35:33.760 versus fraternal? So identical twin concordance within the first few years is only around 50 or 60%.

00:35:40.380 But by adulthood, what is it? By adulthood, it's approaching 80%. So there's a big cohort that Dr.

00:35:45.720 George Eisenbar studied many years ago. So over time, those are the people, the discordant,

00:35:51.200 genetically identical twins are probably the people who would have, they have diabetes,

00:35:56.500 not otherwise specified, but in fact, they have type one that's, that's cryptic.

00:36:00.320 And the fraternal twins, which would at least allow you to hopefully as much as possible,

00:36:04.180 normalize the environmental similarities, but somewhat reduce the genetic variability or the

00:36:10.140 genetic similarity. Do you have a rough sense of what the concordance is for those by adulthood?

00:36:15.180 I don't know the exact number, but again, I think it's, they're approximately tenfold relative to

00:36:20.760 the general population. Got it. Wow. So let's fast forward a little bit and get back to what we

00:36:28.960 talked about over dinner in Houston a few years ago. Again, it was either three or four years ago,

00:36:33.960 but it was the fall. I remember it was definitely a fall. At that point in time, I had spent very little

00:36:38.940 mental energy thinking about type one diabetes, because obviously I'm not going to take care of somebody

00:36:43.460 with type one diabetes. It's so far outside of my skillset. And so I sort of remembered a few things

00:36:49.080 because in residency, two of my co-residents had type one diabetes, which is, that's pretty unusual

00:36:55.020 in a cohort because the categorical residents in general surgery for our year at Hopkins, there were

00:36:59.960 only seven of us. Right. And two of the seven had type one diabetes. And I, you know, I talked to them

00:37:07.000 a lot about it and I sort of remembered directionally their target was a hemoglobin A1c of about 7.5.

00:37:12.580 And they lived on crackers and orange juice where they're sort of staple foods to manage these

00:37:18.680 bouts of hypoglycemia. And I certainly remember one of my buddies, it wasn't uncommon if in the

00:37:23.560 middle of an operation, one of the OR nurses would have to come and put a little straw behind his mask 0.55

00:37:28.400 and he would have to drink some orange juice to sort of, you know, manage his glucose levels.

00:37:32.760 And then when you and I sat down for dinner, you said, you know, your average patient is walking

00:37:38.380 around at something like 5.7, 5.8, 5.9 hemoglobin A1c. And that immediately caught me off guard because

00:37:43.960 I was like, oh, that seems very different from what I sort of remember. So how did you get from

00:37:51.280 wherever presumably you finished your training and entered your practice? I'm guessing that was

00:37:56.160 sometime in the late eighties, early nineties. I graduated from fellowship in, in 2000.

00:38:01.080 Oh, okay. So over the last call it 15, 18 years, how did you evolve your thinking around targets

00:38:08.760 for where the hemoglobin A1c, and I guess I should take a step back and also explain for someone what

00:38:13.260 the hemoglobin A1c is since I'm so critical of it in general, by the way, you've probably heard me

00:38:17.500 rail on the futility of it. The hemoglobin A1c of course measures the amount of glycosylation or how

00:38:23.680 much glucose is sticking to hemoglobin. And if we take the assumption that hemoglobin has a relatively

00:38:29.660 finite fixed, predictable life before the red blood cells themselves get broken down, we can

00:38:36.660 measure the hemoglobin A1c and then impute the average blood glucose. And that's of course the

00:38:43.320 exact opposite of what we would do with CGM, where you are actually measuring the average blood glucose

00:38:48.140 and imputing what the hemoglobin A1c would be. So the diagnosis of type two diabetes is made today

00:38:55.700 solely on the hemoglobin A1c being greater than 6.5%. It no longer relies on the insulin or glucose

00:39:03.920 level during an oral glucose challenge. Remind me, a hemoglobin A1c of 7.5 corresponds to an average

00:39:10.300 blood glucose of approximately what, 160? Yeah, I think that's right. We'll post the table to it. It

00:39:15.780 could be 150, but it's directionally in that ballpark, right? Right. So for someone listening to this

00:39:21.000 thinking, God, that seems a little high. Why was that chosen as such a goal? Okay. So we should

00:39:27.040 first just talk about the DCCT, which is the so-called diabetes control complication trial, which

00:39:32.640 after the discovery of insulin by Banting and Best in 1922. In Toronto, right? In Toronto. That's right.

00:39:40.880 So we're coming up on the 100th anniversary of the DCCT. Of the discovery of insulin. Can we detour for

00:39:47.620 two seconds and explain how Banting and Best did this? Because it's really one of those beautiful

00:39:52.000 stories in medicine. Yeah, it's so amazing. And there's a book, by the way, which is by Michael

00:39:56.620 Bliss called The Discovery of Insulin. And if anybody has a vague interest in type one diabetes, they

00:40:01.140 should run and buy this because it's just so amazing and bring tears to your eyes. Prior to Banting and

00:40:06.280 Best, what happened to children with type one diabetes? Children or adults. So at least in the 19th

00:40:12.620 century, nothing. So these people would present with weight loss and history of drinking and

00:40:18.000 urinating quite a bit. And over time, they had to manage their expectations and then they would

00:40:23.020 expire. And it was, I assume, usually from complications of diabetic ketoacidosis.

00:40:26.860 That's right. Yeah. And so it would happen in a matter of weeks, some cases days, in some cases

00:40:32.000 months. Then along came Allen. And Allen created this amazing so-called Allen diet. And that was

00:40:40.900 essentially a low calorie, high fat nutritional regimen that had almost no enriched carbohydrates.

00:40:48.100 And if you use that approach, they were able to extend life to a few months, in some cases,

00:40:53.580 a few years. But it required incredible precision. And Dr. Joslyn and others had adopted this approach

00:41:01.460 in a desperate attempt to extend life. But other than that, there was nothing you could do and people

00:41:06.040 just wasted away. So it was one of these things that people feared greatly because if your kid

00:41:11.960 got type one diabetes, that was it. You were going to watch them disappear.

00:41:15.900 And you said this earlier, and I didn't interject, but it's probably worth reiterating. You talked about

00:41:20.760 how important insulin is. You use the term catabolic, but in contrast to its role, which is as such an

00:41:27.540 anabolic gatekeeper. And therefore, in the absence of it, you enter this catabolic state. So

00:41:32.140 talk to me about what you mean by the anabolism of insulin.

00:41:36.700 Ultimately, insulin is a key factor that's regulating virtually every cell in the body. And

00:41:41.980 it's essentially deciding whether a cell is in a state of feast or famine. And so if you're in feast,

00:41:50.060 it's time to take nutrients and park them in the cell. And if it's famine, hey, let's mobilize some

00:41:55.120 nutrients and make them available to keep the organism alive. And so it's a master regulator of your state of

00:42:01.440 metabolism. And so if you take a normal non-diabetic human and you fast them for 24 hours, you discover

00:42:08.760 that they're circulating insulin values, especially of a healthy young adult, are exquisitely low.

00:42:13.440 But if you then take that same person, you give them a 75-gram glucola challenge, you can get their

00:42:19.220 insulin values to exquisitely high levels. So it's a dynamic hormone. It's changing rapidly in response

00:42:25.440 to meals and everything else. And it's helping you to sort of take these nutrients and park them away.

00:42:30.580 Now in the fat cell, I think most people have a vague notion of what's going on. There's basically

00:42:36.900 one door into a fat cell of esterification or re-esterification, and then there's the door out,

00:42:41.560 which is lipolysis. And insulin acts very prominently at both of those. So free fatty

00:42:47.140 acids and triglycerides are highly regulated by insulin at both ends of that adipose cell.

00:42:53.360 What is insulin doing at a muscle cell?

00:42:54.940 It's promoting glucose uptake into skeletal muscle. And it's a dynamic process that involves

00:43:00.780 GLUT4, these glucose transporters. And essentially the glucose transporters are normally sitting in

00:43:06.160 the cytosol and they translocate to the cell wall. And now all of a sudden the muscle cell becomes

00:43:11.740 permeable to glucose and glucose is pumped in.

00:43:15.140 So GLUT4 is a passive transporter. As long as it's going across the cell membrane, glucose enters

00:43:21.140 freely. The active part of that is the insulin signal that tells the GLUT4 transporter to go from

00:43:26.420 the cytosol across the membrane.

00:43:28.840 Right.

00:43:29.380 And then, so taking the other extreme of that, which is when someone is fasting or starving,

00:43:35.080 which is effectively what's happening in type 1 diabetes at the cellular level, they don't have

00:43:39.340 insulin. How does the lack of insulin promote the breakdown of actual muscle? So this negative

00:43:45.260 nitrogen balance could, because I'm assuming that while a lot of that muscle that breaks down

00:43:49.960 is used for gluconeogenesis, wouldn't that patient, if you're measuring urinary nitrogen levels,

00:43:55.040 also be in a profound negative nitrogen balance?

00:43:57.340 Yeah. Every step of protein metabolism, many of the major steps are influenced by insulin as well.

00:44:03.700 So you're either making proteins or you're breaking them down. It also influences the glycogen that's in

00:44:09.300 skeletal muscle and heart and kidney. So in the presence of insulin, you're putting more glycogen into

00:44:15.420 skeletal muscle or in the absence, you're going to, you're, you're going to break it down or also

00:44:19.660 with just work.

00:44:20.760 Boy, it's a thin razor's edge, that hormone, too much of it, really bad stuff, too little of it,

00:44:26.660 really bad stuff. As long as you can control it, it's a pretty wonderful hormone.

00:44:31.120 And it, because it's so fundamental to eukaryotic cells, it seems to also influence things you would

00:44:37.380 never even imagine. Like for instance, lifespan and aging. And so it's re it's really a deep and

00:44:43.500 fundamental signal around where cells are.

00:44:46.060 Do you have a great sense of how insulin relates to IGF? It certainly plays a role in

00:44:51.700 the inverse relationship between some of the binding proteins. So we know that

00:44:55.800 as insulin level goes up, sex hormone binding globule goes down and therefore high levels of

00:45:01.140 insulin will all things equal reduce the level of circulating androgens because more of them are

00:45:06.200 being bound up. What's the relationship between insulin and the binding protein for IGF, IGF binding

00:45:11.940 proteins, two, three, et cetera?

00:45:14.140 I'll probably get this wrong, but I think it's IGF-1 can go down in the presence of high levels

00:45:20.820 of circulating insulin acutely. And it's one way to be able to diagnose a very short-term

00:45:27.900 hyperinsulinemia. This is something that we often struggle with in pediatrics. We bizarrely even have

00:45:33.220 some people who will take way too much insulin trying to pretend that they have hypoglycemia by some

00:45:38.960 measure. And then we go and measure in the blood and you can actually measure for IGF-BP-1

00:45:42.800 and it can be down.

00:45:45.120 So the BP-1 goes down. So in many ways, IGF would go up then. If the binding protein goes down,

00:45:51.100 the free IGF should be going up with insulin, right?

00:45:53.360 Assuming I remember this correctly.

00:45:54.980 Well, that's sort of what I see clinically, right? The more hyperinsulinemic a patient is all

00:45:58.800 things equal, the higher you tend to see their IGF. And that's always been my assumption as to why,

00:46:03.960 just based on the parallel analogy with the... I'm ashamed to say I'm too lazy to have spent the time

00:46:08.900 looking this up.

00:46:09.580 Well, the other thing I would just point out is that insulin and IGF-1 really are intensely

00:46:15.280 overlapping in the way they work. And insulin does not bind as a monomeric hormone. It's actually a

00:46:21.680 dimer and it can be a heterodimer. So you can have two insulins or you can have an insulin and IGF-1

00:46:27.700 or you can have two IGF-1s and they can bind to receptors. And there are even heteroreceptors where

00:46:33.480 it's an insulin receptor and an IGF-1 receptor, each bound by their cognate ligands. Derek Leroyth

00:46:39.520 at Mount Sinai has done this kind of work for years. And when you block these heterodimeric

00:46:44.400 receptors, you can completely alter metabolism, implying that the role of IGF-1 on glucose homeostasis

00:46:51.480 is really quite powerful.

00:46:53.680 Yeah. And we probably won't get to it today, but there's also a IGF and its relationship to

00:46:58.740 growth hormone is actually quite interesting when it comes to diseases because it seems to have very

00:47:03.400 different effects on different diseases. There are certain diseases that seem to go up in risk as

00:47:08.040 IGF goes up and there are others that seem to go down. Whereas insulin seems a little less confusing,

00:47:13.280 but we'll obviously come to that. So in the late, you know, 1910s when Banting and Best are working

00:47:19.800 with dogs, do they know that they're trying to isolate? Like, do they have a sense of where to

00:47:24.180 look for this hormone? 0.99

00:47:25.640 Yeah. So Banting has an idea. Actually, the story of how he comes up with the idea is amazing and

00:47:31.220 really surprising. So he's a surgeon in a dead-end practice in London and-

00:47:36.860 This is London, Ontario.

00:47:37.740 London, Ontario. And he can't make his mortgage. He's a veteran of the World War. At that time,

00:47:44.420 there had only been one. And he's trying to get this general surgery practice going, but he's bored.

00:47:49.660 And he doesn't have that many patients. And so he's a reader and he's thinking about ways

00:47:54.020 to try to contribute. And he reads in this journal article, which describes how people who have stones

00:48:01.820 in the pancreas end up with these strange islet-like structures or islands of cells.

00:48:10.140 And so essentially what happens is the pancreas, for whatever reason, if it's obstructed by a stone

00:48:15.320 and it has a blockage, what you end up with in the residual after the pancreatitis occurs are these

00:48:22.980 weird round cellular structures. And so it had been known from a couple other scientists that

00:48:28.720 there might be something in there that was involved in diabetes. So he comes up with the idea of, wow,

00:48:34.980 maybe what I could do is block the outflow of the exocrine pancreas, which makes these digestive

00:48:40.700 enzymes, force the pancreas to eat itself alive. And then I will take this residual muck that comes 0.86

00:48:47.580 from the pancreas and try to purify whatever is within it to see if that is some sort of medicine

00:48:54.600 that might be useful for type 1 diabetes. It's a genius moment that comes to him in the middle of

00:49:00.000 the night. He writes it down. Michael Bliss actually found the notebook page from his bed stand where

00:49:06.540 he wrote this. And so then he goes to University of Toronto, which is this very sophisticated place.

00:49:12.660 And he meets and he knocks on doors and he finds this guy who's a famous professor of metabolism.

00:49:18.100 And he says, I got this incredible idea. I want to isolate this thing that must be in the pancreas.

00:49:23.460 That would be the treatment for type 1 diabetes. And the guy's like, you're out of your mind.

00:49:29.060 He's like, no, no, no, it's a really cool idea. Let me tell you about it. And what happened is

00:49:32.700 this professor is about to go on break for the summer and to go off to Scotland. And so he leaves

00:49:41.260 Banting with a very short lease. He says, look, I'll allow you to do a few experiments

00:49:46.520 and I can give you a few resources, but you basically got to finish your experiments by the

00:49:51.340 end of the summer. And so Banting is there. And then he essentially gets a medical student who

00:49:57.080 has the dumb luck to draw this crazy experiment with Banting. And that's Charles Best. So there they 0.87

00:50:04.340 are in this cramped, hot laboratory in Toronto. And amazingly, they are able to isolate this

00:50:12.460 substance that has the ability to lower blood glucose in dogs.

00:50:17.240 And they're doing this in dogs as well?

00:50:19.260 Yes, that's right. So what they do is they carry out this intervention. The pancreas eats itself

00:50:24.540 alive. They get this muck and they are able to isolate some crude extract and then take it and put

00:50:30.340 it back into a dog that they've removed the entire pancreas of and lower the blood glucose. It's like

00:50:35.880 this incredible moment. I mean, and this is 1922. What's even more extraordinary is that they're able

00:50:42.500 to take this observation and then they eventually fairly quickly do an experiment where they lower the

00:50:49.880 blood glucose of a medical student who has type 1 diabetes with this super crude extract. So they're

00:50:56.080 able to prove that it works in man. And so then they write the paper and the paper gets rejected.

00:51:03.400 You know, and then they're able to submit it to a Canadian journal and it gets accepted.

00:51:08.940 That low Canadian bar. I love it, man. That's like been the secret to any success I've had.

00:51:13.800 And they ultimately won the Nobel Prize. It's amazing. And if you think about 1922.

00:51:19.240 And by the way, it's sort of one of like every year there is a Nobel Prize awarded in medicine or

00:51:24.580 physiology, I believe is the designation. There are a handful of them that stand out in a way that

00:51:30.340 can't be described. And this is one of them. The Banting and Best Nobel Prize is, you know,

00:51:36.080 one of a handful of places where the trajectory of clinical practice was altered dramatically.

00:51:43.000 You know, I think just by virtue of the fact that we're both physicians, we can be,

00:51:48.040 we can feel proud by association that our profession was able to do something so remarkable.

00:51:52.660 How long was it before recombinant insulin was used? Because after Banting and Best,

00:51:59.260 they were still using insulin from animals.

00:52:01.500 Yeah. Purified from pigs. It's amazing. It went on for years and years and they would take these,

00:52:05.200 these giant vats and it was either from beef or from pigs.

00:52:08.460 It wasn't until the eighties, if I recall, that they figured out how to make insulin inside of E.

00:52:12.940 coli once, you know, Paul Berg.

00:52:15.080 Genentech.

00:52:15.640 Yeah.

00:52:15.900 Yeah. So there's an amazing article about the molecular biology of modern insulin.

00:52:21.460 And, you know, that was basically a race and it was a bunch of. 0.95

00:52:25.900 It really was. It's a biotech arms race to figure out. And it's some of the greatest stories in science

00:52:32.280 are of that arms race of the eighties.

00:52:34.560 They knew that human insulin could be incredibly useful for this population. And part of the reason

00:52:40.400 is if you're taking a foreign product like pig insulin, you can develop antibodies. And in some

00:52:45.100 cases, those antibodies can be blocking to the point where insulin, you start to have to increase

00:52:50.740 the doses greater and greater. So human insulin was known to be a goal and they were eventually

00:52:55.860 able to produce it. Genentech did this, they sold the rights, and now there are multiple modern

00:53:01.480 human insulins on the market. Not to get into a controversial topic, but can somebody explain to

00:53:06.780 me, and you can, you can opt out of this question if it's too politically sensitive, but can somebody

00:53:11.440 explain to me why the hell recombinant insulin is still so expensive? Well, what's interesting about

00:53:16.340 recombinant insulin is it used to be cheap. So back when I was a fellow, a vial of insulin, which is a

00:53:20.980 thousand units, cost about 25 bucks. And that would last, that could last a well-managed patient for a

00:53:27.560 month. A month or more, right. And today, what is a thousand units?

00:53:31.480 Of recombinant insulin. 300 plus, maybe $400 US. Right. Now I don't know a lot about economics

00:53:38.060 and I kind of forgot my tables of inflation, but I feel like that has outpaced inflation a little bit.

00:53:44.860 So how is it that we have more competitors in the space, better technology today, and we've watched

00:53:51.280 a log fold increase in the price of this stuff that must be driving a lot of people nuts, including

00:53:57.600 those who can't, who don't have insurance. And frankly, the insurance companies themselves that

00:54:01.760 have to pay this. It's a disaster on many levels. It's really a disaster for people who are on the

00:54:08.260 margins of society, but want modern care and are willing to pay for it, but they can't get insulin

00:54:15.060 at the same price that for instance, you could, if you walked into a pharmacy in Spain, you could buy a

00:54:21.040 vial of insulin for 25 bucks there. It's three or $400 here. The difference, unfortunately, is that

00:54:27.700 the federal government is a, how shall I say it, a promiscuous customer. So they are thus far have

00:54:34.700 not been able to drive the price down because they're willing to essentially pay whatever the

00:54:40.980 sellers charge. And that's a big problem. And it speaks to the lack of market influence

00:54:45.780 in things like insulin. So now let's go back to where I distracted you from a long time ago,

00:54:51.660 which is you finish your fellowship. You're a card carrying pediatric endocrinologist who knows how

00:54:57.500 to target high-ish levels. And we were about to talk about the trial that explained the complications

00:55:04.980 that basically led to the why we care about this. Okay. So if the discovery of insulin is like the first

00:55:11.140 real modern chapter in type 1 diabetes, then the diabetes control complication trial is chapter 2.

00:55:18.920 And it's based on a brilliant and very clear idea, which goes, if people with type 1 diabetes are at

00:55:25.600 risk for terrible complications, and we haven't really said this, but they're at risk for all the

00:55:29.720 complications that people with type 2 diabetes get, cardiovascular disease and kidney disease and

00:55:36.360 diabetic retinopathy and blindness and limb amputations. Well, actually, I'm glad you brought

00:55:41.960 that up because I've been a little delinquent in my job as host and taking things for granted.

00:55:47.640 Can you explain why people with type 1 or type 2 diabetes have so many of these complications that

00:55:54.300 we refer to, I guess, loosely as microvascular complications? And we'll get to the macrovascular,

00:55:59.160 of course, but just start with those ones you mentioned. Why the amputations? Why the blindness?

00:56:03.360 The blindness is from abnormal vessel growth. And it appears that there are these weird

00:56:08.700 glycations that can occur. And there's a series of damages that happen in the vascular endothelium.

00:56:15.800 And then there's new vessels that are created, and the new vessels are fragile,

00:56:19.280 and they can ultimately break and bleed. And the kidney, it's a very different set of mechanisms,

00:56:25.920 but basically very high glucose can alter the epithelia of the kidney that's involved in filtering

00:56:32.500 blood and making urine. And then why the amputations of toes and things like that?

00:56:36.880 Yeah. So that's altered healing from a couple of things. So one is that there's something called

00:56:40.960 diabetic neuropathy, where the nerves simply don't work well. And exposed to high glucose,

00:56:47.200 they lose their responsiveness. And so sometimes people will, for instance, have a wound in their

00:56:54.160 foot, but they're not able to actually realize that they have that wound. And so then they also have

00:56:59.140 abnormal healing because the microvasculature isn't behaving appropriately. And the sum total can be

00:57:06.040 very bad diabetic wounds that fail to get better with antibiotics and ultimately result in amputation.

00:57:11.560 And this is like the most common cause of preventable amputation in the United States. It's also the

00:57:16.960 most common cause of preventable blindness in the United States, and probably also renal failure.

00:57:22.520 I was just about to say, it's got to be the leading cause of kidney disease.

00:57:25.840 So the relationship between the average blood glucose, which we proxy by this hemoglobin A1c,

00:57:33.140 and these complications you mentioned, that's a monotonically increasing relationship, correct?

00:57:39.160 Pretty much, yes. So the DCCT really figured this out. So at the time they designed this trial,

00:57:45.600 they're thinking, well, all this terrible stuff is happening to these people with type 1 diabetes.

00:57:49.020 And so there was a group of people who thought, well, it's just genetics. You know, it's obvious

00:57:54.080 because we have two patients and one who has pretty good control. The other one has pretty

00:57:58.060 good control, but one gets complications. What were the years of enrollment in the DCCT?

00:58:02.200 So this is in the late 80s, actually mid 80s, and it was reported out of the New England Journal,

00:58:06.460 I believe in 1993.

00:58:08.320 And prior to that, you're saying it wasn't obvious that the differences in these complications

00:58:13.900 could be explained by the differences in their average blood glucose.

00:58:17.000 That's right. And there were endocrinologists who believed quite strongly that it was foolish 0.95

00:58:21.860 and perhaps even dangerous to carry out a clinical trial to test the basic idea. So the core idea is,

00:58:28.820 does tight control, i.e. near normal blood glucoses, confer protection from diabetes complications?

00:58:36.240 And the primary outcome measure was, were microvascular. It was kidney and eye.

00:58:41.620 And the reason that these physicians who we can look back at now, and obviously with the benefit of

00:58:47.360 so much hindsight, sort of consider that their views were silly, they weren't bad doctors, right?

00:58:52.060 They had something to be afraid of on the other end of this, right?

00:58:54.540 They were concerned about hypoglycemia.

00:58:56.380 So talk about that for a moment. That's it. Because there's got to be a really bad boogeyman

00:59:00.320 on the other side of this for people to be so worked up about letting glucoses get really high.

00:59:05.160 So the issue with type 1 diabetes is, we think that high blood glucoses are conferring risk of

00:59:09.660 complications, but the problem is you can't just throw in a bunch of insulin and then normalize

00:59:15.220 glucose because you will cause somebody to go low. And with type 1 diabetes, the crux of the problem is

00:59:20.840 the volatility in blood glucose. So it's not just so simple that you can figure out the amount of

00:59:27.000 carbohydrates you're going to eat and take some insulin and your blood sugar is going to be normal.

00:59:31.560 Because even if you eat the exact same thing every single day, as anyone knows who has type 1 diabetes,

00:59:37.420 and I only know this because of friends who have it, your level of stress, how much sleep you had

00:59:42.360 last night, how much exercise you've done or not done, right? Can even in the presence of the

00:59:47.440 identical food stuff produce completely different insulin requirements.

00:59:52.160 And the other things in food like fat and protein also alter gastric emptying. And the end result is

00:59:58.460 a tremendous amount of slop and variance in the system. And so we in the medical field often have

01:00:05.540 this fantasy that, well, we can just come up with some sort of formula for, here's your insulin to carb

01:00:11.600 ratio. And so I tell you that, and then you're supposed to follow it. And then if you come back and your

01:00:17.060 blood glucoses aren't exactly perfect, I, the physician would say, well, you obviously didn't follow it.

01:00:21.980 You're, you know, they use this terrible term in medicine, non-compliant, which means you, you didn't

01:00:27.260 comply with what I told you to do. It's grotesque.

01:00:31.620 In no format or no arena is it more ridiculous than this one when you realize how difficult this is. 0.57

01:00:38.840 And as I learned from a patient of mine, actually, this was actually probably the first person I was

01:00:43.100 ever involved in the care of as a medical student. And we became very good friends. He ultimately died

01:00:48.420 from colon cancer, but he was a pilot. And after, during one of his bouts of, you know,

01:00:54.420 sort of remission, he took me flying and he actually let me fly. Once we got up, he, you know,

01:00:59.740 he took his hands off the stick. I had a stick and I was flying. And he said, you got to look at the

01:01:04.320 horizon. Don't look at the instruments. Because if you look at the instruments, you're going to

01:01:08.040 constantly be chasing them and you're getting to this pilot induced oscillation. And obviously that's

01:01:13.260 called instrument flying, which you learn to do way down the line. But as a total rookie,

01:01:16.920 you have to be able to fly off the horizon. And it was amazing to me how much easier it was to fly

01:01:21.780 that way. And in some ways that's what happens when patients or physicians would get frustrated

01:01:26.840 with this inability to manage this as you, you tend to go too high, then too low, then too high,

01:01:32.120 then too low. And then you are vacillating wildly in these glucose levels.

01:01:36.600 And again, yeah, you see these huge sine wave excursions and, and some of them can be incredibly 0.79

01:01:41.460 volatile. So I want to go back to the DCCT. So they have this idea that they're going to somehow

01:01:45.420 normalize blood glucose. But then there are some people, physicians who say, well, that's a terrible

01:01:49.960 idea because you can't actually normalize blood glucose because you're going to make a bunch of

01:01:53.860 people hypoglycemic. So there's a group of people who are arguing against the DCCT. There's another

01:01:58.280 people, including Elliot Joslin, who's saying, no, we need to understand if we can reduce the risk of

01:02:04.300 long-term complications. So they carry out this trial. It's intended to be a 10-year clinical trial.

01:02:09.040 And they, it's 1,441 patients. They're relatively recent onset type 1 diabetes. They're largely

01:02:16.720 late teen and adults. And it's a very difficult trial to do because there were no established

01:02:23.340 best practices to achieve neuronal blood glucose at the time. So the average hemoglobin A1c for the

01:02:28.880 population is around 9% at the time. That was just where they were.

01:02:32.960 That's the lowest glucose you could have to not have hypoglycemic events, right?

01:02:36.380 Yeah. And sort of, that's where people were in order to not dump massive amounts of glucose in

01:02:41.660 their urine and sort of look okay. And in some cases, people were taking insulin once a day.

01:02:47.020 Quite often people were taking insulin once a day.

01:02:49.620 And someone who's got a hemoglobin A1c of 9% is not going to live a normal life. Their life

01:02:53.960 expectancy is truncated dramatically and their quality of life is going to be,

01:02:58.940 even by their forties, they're going to be quite compromised.

01:03:01.220 Yeah. So we know now from Swedish studies that somebody with an average hemoglobin A1c has about

01:03:07.860 a- Of what? Of 9?

01:03:09.380 Of 9%.

01:03:10.000 9%.

01:03:10.400 Has approximately a six-fold increased risk of death and a six-fold increased risk of

01:03:17.620 cardiovascular disease. And so that's shocking. And again, these are healthy people who are doing

01:03:23.100 their best, but the system just didn't work. So in the DCCT, they say, well, we want to normalize

01:03:28.540 blood glucoses in the population and try to get them from nine all the way down to the non-diabetic

01:03:33.640 range, somewhere under six. And they put together all of these centers. And what they did was they

01:03:38.800 mandated conference calls. And on the conference calls, they shared best practices. They crowdsourced

01:03:44.520 it, if you will. And so they didn't have any idea how they were going to achieve their goals,

01:03:49.180 but they ended up ultimately learning how to do it. And it involved phone calls and it involved

01:03:55.960 frequent visits and cajoling. And they're able to reduce the A1c from 9% to 7% in the population.

01:04:03.280 They intend to do this for a decade. There's a safety and monitoring board that's following the

01:04:09.100 curves. The primary outcome measure is diabetic retinopathy and diabetic nephropathy. And the

01:04:14.780 safety and monitoring board sees a huge difference in between the control and the intervention group.

01:04:19.340 And they stopped the trial. And the reason is they felt that this was seven years in,

01:04:26.500 and they felt that they- It was now unethical in the reverse.

01:04:29.760 Yes. It was unethical to deprive the general population of what they had learned. So they

01:04:35.520 stopped the trial. There's a giant talk at the American Diabetes Association that describes these

01:04:40.340 immense differences. There's a paper that's in the New England Journal of Medicine. It's amazing,

01:04:45.780 wonderful paper. And it changes type 1 diabetes forever because it says that we can begin to

01:04:52.080 reduce complications. And the work that we do is incredibly important for this patient population.

01:04:58.860 In some ways, it makes the field of modern type 1 diabetes care because it says there really is

01:05:05.840 something you can do other than hold somebody's hand and look for complications. But on the other

01:05:10.900 hand, it actually breaks the field because we don't have the apparatus to routinely achieve

01:05:16.580 the kinds of outcomes that the NIH was able to do because they had near infinite resources.

01:05:21.620 They spent over $100 million for a relatively small number of patients.

01:05:25.040 $100 million for 1,400 patients over seven years?

01:05:28.440 Yeah.

01:05:29.080 That's a staggering sum of money. Where did it go? I mean, and I'm not being critical of that. I'm just

01:05:33.560 saying like, what was so expensive about that study?

01:05:36.340 Yeah. Coordinating centers and buying people's time to be able to care for them, to call them,

01:05:41.360 cajole them. There's a lot of work that was done.

01:05:44.420 Wow. I did not realize it was that expensive. And that's 30 years ago's dollars.

01:05:49.140 Yeah. I could have it wrong. We'll have to look this up.

01:05:51.280 Okay. Yeah. We've got the team to look that up. But I'm guessing directionally,

01:05:55.260 you're right that it was a costly study.

01:05:57.000 It was a vast amount of money. Yeah.

01:05:58.240 Was there a sizable or noticeable increase in the hypoglycemic events with that tighter control

01:06:04.000 from nine to seven? Yes. There were many more hypoglycemic events and there was weight gain.

01:06:08.920 And so when they jack up the insulin, people found themselves eating more carbs and eating

01:06:13.300 out of the hole. And over time- Was that actually the case that they know that the people who gained

01:06:18.160 more weight consumed more of one type of food than another? No, not necessarily. But practically,

01:06:24.880 if your blood glucose is low, the only thing that's going to get your glucose back up is carbohydrates.

01:06:28.960 Oh, so I see. So it's, they were chasing carbs when they were overshooting on insulin.

01:06:33.920 Right. And this is one of the continual problems with type 1 diabetes is if you jack up

01:06:38.200 the insulin and your blood glucose is low, there's only one thing you can do to get your glucose back

01:06:43.360 up. I guess you could go out and you could exercise intensely, but it won't happen fast enough.

01:06:48.260 So acute carbohydrate is essential.

01:06:50.620 So we don't really know, not to get off on a tangent, we don't know if those patients gained weight

01:06:54.840 because of the hyperinsulinemia per se as a primary driver, or if it was sort of not necessarily a root

01:07:03.420 cause, but more of a tangential driver that ultimately forced more of this sort of hyperphagic

01:07:08.520 response indirectly to compensate. You're asking a really important question, which is,

01:07:12.940 is the weight gain in type 1 diabetes absolutely obligate? Is it coincident with type control?

01:07:19.780 And it was argued by many that the only way towards type control would be weight gain.

01:07:26.300 And by the way, in the years since the DCCT, what we've seen is a huge number of people with type

01:07:31.020 1 diabetes have gained weight, in some cases uncontrollably. And that is a really serious

01:07:36.960 problem. And moreover, in the, the people who have gained the most weight have had poor

01:07:42.280 cardiovascular outcomes. There's a weak association. So I want to just talk about the bigger picture of

01:07:47.080 the DCCT. So the, the trial only happened for seven years and then the cohorts merged back to

01:07:53.940 the same blood glucose. And now they're followed for 25, 30 years.

01:07:58.060 Wait a second. When they merged, did they merge to seven or merge to nine?

01:08:02.220 Yeah. They merged to eight. So what happened was the people who had hemoglobin A1Cs of nine learned

01:08:07.940 the benefits of tighter control and they did their best.

01:08:11.000 And the people at seven, no longer under the, you know, grip of the best in class care drifted up

01:08:17.300 to eight. That's right.

01:08:18.420 Eight became where people landed.

01:08:20.100 Eight became the new normal with a significant standard deviation within it. Some people just

01:08:25.240 did better or worse. And then you follow these people over the subsequent 23 plus years. And it

01:08:31.860 turns out that there are immense differences in, in major kidney disease and in cardiovascular disease

01:08:38.640 and even death. So the people who at least for seven years had tighter control versus less tight

01:08:45.300 control, but then for 23 years were identical to their peers, still retained some benefit from the

01:08:51.380 seven years. Yes. 30 years earlier in cardiovascular disease, which is amazing and death. So that,

01:08:57.520 what kind of hazard ratios do you recall? I'm not sure I'll be able to remember it off the top of my

01:09:02.300 head. We'll link to all of this stuff. At least in the primary outcome measure for the,

01:09:06.700 for the cardiovascular trial, it's approximately a third less of the primary outcome measure.

01:09:12.280 So that's really exciting. And it suggests that if you were able to routinely achieve blood

01:09:17.080 glucoses in the near normal range, you could drop your risk of cardiovascular disease

01:09:20.880 accordingly to something that was approaching a normal, healthy, non-diabetic person. That's pretty

01:09:26.300 amazing, but it also puts a lot of pressure on us as a field.

01:09:29.360 So if you were to have a device, a magical device that could measure someone's glucose in real time,

01:09:36.480 and you were to put this on somebody 30 years ago, back when the standard of care was,

01:09:42.660 you know, minimize the hypoglycemic events, we're going to tolerate a hemoglobin A1C of 9%.

01:09:48.120 Can you estimate what the standard deviation of glucose would have been on that measurement? Do

01:09:52.900 you have a sense of how high that could have been?

01:09:54.980 Yeah, because I put continuous glucose monitors on teenagers who don't listen to me.

01:09:59.000 Okay. So talk to me about some of those numbers.

01:10:00.780 Yeah. So you might see blood glucoses of around, in some cases, 180 or 200 with a standard deviation

01:10:07.740 of 100. And so again, poorly controlled type. You know, my standard for myself, right? Which is

01:10:13.920 under 20, 90 plus or minus if I'm fasting or on a, you know, sort of ketogenic diet,

01:10:19.820 I'd like to see a plus or minus less than 10. And when I'm just eating ad libitum, it's sort of less

01:10:24.440 than 15. A hundred's amazing. A hundred means you're bouncing around nonstop and you're never

01:10:29.980 normal. And you feel so bad for these patients because even if they intermittently decide,

01:10:35.280 okay, today's the day I want to get this right. Like you just can't chase that down. That is so

01:10:39.700 difficult. I don't even know. It's hard for me to physiologically even think about what's

01:10:43.400 happening in those patients. And so that kind of volatility has just driven everybody insane

01:10:47.900 because they think like, there's got to be a better way. There's got to be a better way. And you know,

01:10:51.080 anybody who lives with type one diabetes, they're constantly being asked, okay, well,

01:10:55.460 what about a cure? I hesitate to even use that word, but like some sort of definitive biological

01:11:01.060 therapy. And, you know, and we've heard all along about the possibility of stem cells or some sort of

01:11:08.120 maybe an auto loop pancreas that would measure your glucose and administer insulin, or maybe some sort

01:11:14.180 of beta cell regenerative therapy or something combined with something that shuts down the autoimmune

01:11:19.780 system. So those three concepts are constantly trotted out as potential avenues. But from what

01:11:26.620 I can tell as a scientist, many of those things are years and years and years away, decades, maybe

01:11:31.480 even centuries. And so the problem is like, what do you do now to support people who live with type

01:11:37.140 one diabetes? How do you reduce that volatility?

01:11:39.380 So I remember reading a paper a few years ago, it was actually right after you and I met that I got

01:11:44.580 more curious about this. And especially then once I started wearing a CGM, which was shortly after

01:11:49.740 that, and I've basically never taken it off. I started to figure, like, I started to think, okay,

01:11:55.200 well, from my angle, it's how do you use these in people who don't have diabetes to drive better

01:12:00.600 health, right? And one of the questions I wanted to understand was, was there any evidence that

01:12:05.400 variability in blood glucose per se beyond the average? So if you have someone whose average blood

01:12:09.980 glucose is 100 plus or minus 10 or 15 versus someone who's 100 plus or minus 30, is there a

01:12:16.480 difference? Well, the hemoglobin A1c would give you the same answer, and even the average glucose

01:12:20.900 does. And I believe I found some data that suggested that cognition was impaired during

01:12:26.460 developmental years by higher variability. And I think it was even independent of absolute glucose

01:12:31.980 level. And that was very frightening to me, not individually because I was out of that loop,

01:12:37.000 but thinking about kids with type 1 diabetes who now have yet another complication to be concerned

01:12:43.000 with, which is cognitive impairment, if the variability of glucose is too great.

01:12:47.440 It's well known that diabetes complications are increased with increasing A1c's. A kind of

01:12:53.580 surprising piece of this is very short exposures to high blood sugars can completely change gene

01:13:00.460 expression in a whole bunch of genes like endothelial cells. So there's a school of thought that says

01:13:06.400 that diabetes complications are really around the exposures to these weird toxic levels of blood

01:13:13.340 glucose that can occur.

01:13:14.900 And what do you think toxic is? How high?

01:13:16.760 Well, you know, they always do it experimentally by turning everything up to 11. So you never really

01:13:21.140 know. But what you can say is it is surprising and weird that a short, almost evanescent exposure to

01:13:28.780 a very high blood glucose could change the way genes behave in cells in a dish. And so that suggests to

01:13:38.000 me that diabetes complications may not be so simple as a sort of on-off in between glucose and hemoglobin

01:13:45.800 or glucose and some other protein. You could have permanent profound changes in the way a cell behaves

01:13:52.380 based on intermittent exposures to very high blood glucose. So whether that's volatility or it's an absolute

01:13:58.860 level, we don't know.

01:13:59.320 That's an absolute peak. Wow.

01:14:00.720 But it still puts... 0.99

01:14:02.540 That's just like nature to fuck with us like that. I mean, to have to deal with some epigenetic change in 0.98

01:14:08.140 response to a glucose. I mean, that's so unfair.

01:14:10.760 But in the end, it shows you that we had a very superficial understanding of diabetes complications years ago.

01:14:16.740 And again, now there's a whole rich field that's devoted to this idea of these windows of hyperglycemia

01:14:22.720 permanently altering these important cells that ultimately contribute to diabetes complications.

01:14:28.560 You're making me question my elaborate cheat meals sometimes. Just admit this now to you and

01:14:35.960 everybody else. A couple of days ago, I was like, you know, I've been so good for so long. I've been

01:14:40.820 doing my this, that, and everything. And there were just a handful of junk foods I can't get enough of.

01:14:46.040 And one of them is mini wheats. So I decided to have a box of mini wheats.

01:14:50.560 That's not a typo. It's not a bowl of mini wheats, a box of mini wheats.

01:14:54.100 Right.

01:14:54.300 Meaning if you have a big enough bowl, you can put the box in the bowl. So I had a box of mini

01:14:57.840 wheats in a bowl. And boy, if you really want to know why that's not a good idea, just wear a CGM.

01:15:05.300 And not to mention the splitting headache I had like an hour later.

01:15:08.840 Yeah, I get headaches.

01:15:09.940 I mean, I thought I was going to die. I'd feel better if I had six drinks, I think my head would have hurt less.

01:15:16.040 My blood glucose hit like 170. Now, again, that's in the grand scheme of things,

01:15:21.460 probably not that high for someone who's got type 2 diabetes or type 1 diabetes. But for someone who's

01:15:26.160 not diabetic, that's a pretty staggeringly high. I haven't seen a level that high in years.

01:15:30.500 And you think like, hey, these mini wheats are promoted as like healthy, part of a healthy

01:15:35.160 breakfast.

01:15:36.080 And so imagine now-

01:15:37.360 Of course, no one's eating a box, I hope.

01:15:39.200 But, you know, if you think about this in the context of your kids, imagine dropping your kid

01:15:44.380 off who, if your kid has type 1 diabetes, you drop them off at a birthday party. And there's cake,

01:15:51.260 and there's, you know-

01:15:52.160 Capri Sun.

01:15:52.660 Capri Sun.

01:15:53.380 Yeah.

01:15:53.560 And there's Cheetos. And all this stuff is essentially carbohydrate. And if your child is-

01:16:01.520 And none of them are real carbohydrates. That's the problem, right? It's all like these fake

01:16:05.160 carbohydrates. It's one thing if they were like, all right, we're going to grind up some wheat and,

01:16:10.520 you know, you're going to have a piece of bread, like, which has its own issues. But yeah,

01:16:13.480 these things that you've described are, they're like, they're basically IV infusions of glucose.

01:16:18.560 Let's imagine that you drop your five-year-old off at a birthday party. And you know they have

01:16:23.980 diabetes, and you give them, and they're wearing an insulin pump, and everyone knows what they're

01:16:29.240 supposed to do. And out of sight of all the adults, the kid eats two cupcakes. That's the

01:16:36.340 equivalent of the box of wheat thins, is that what you-

01:16:39.340 No, but it's funny you bring up wheat thins. That's one of my kid's favorite things,

01:16:42.800 is he calls them wheat-ins, but no, it's mini-wheats.

01:16:45.120 Mini-wheats.

01:16:45.500 Mini-wheats, yes.

01:16:46.580 It's just nonsense. It's basically breakfast dessert.

01:16:49.100 So again, this thing was really perplexing for me. And part of, you know, when I really

01:16:54.260 understood how challenging it all was, was spending time with people who live with type 1 diabetes and

01:16:59.760 looking and seeing what they do. And I didn't really get this as an endocrine fellow. And

01:17:05.520 we had this standards feel about, okay, you're going to check your blood sugar before each meal

01:17:10.820 and at bedtime. And then you're going to eat a certain number of meals that involve a certain number

01:17:15.260 of carbohydrates. And we had this fantasy that we're just going to sort of make people live in

01:17:21.980 what I call the diabetes ICU, which is a prescribed number of carbohydrates and timing. And then we're

01:17:28.360 going to sort of work towards making sure that everything was very predictable and that blood

01:17:33.360 glucoses would be near normal much of the time. And I lived in that fantasy world unperturbed by

01:17:40.560 reality until I became close friends with several people who have type 1. And then I began to see

01:17:46.720 what they do. So for instance, my friend Marianne is a pediatric endocrinologist who works at Boston

01:17:53.280 Children's. And one time we were at this restaurant and she's, we were at Bertucci's, you know, so pizza

01:18:00.100 place. And she- And she has type 1 diabetes.

01:18:02.840 She has type 1 diabetes and she's eating a Caesar salad. And I see her picking the croutons out. I'm

01:18:08.900 like, what are you doing? And she goes, oh, Jake, I can't eat these things. I'm like, what are you

01:18:13.660 talking about? She's like, if I eat those, these carbs, my blood sugar will just go loopy. So what I

01:18:18.900 discovered- And let's pause for a moment. You know, all of the molecular biology, you know how many

01:18:24.560 grams of carbs are in that. You've been managing patients through their food with insulin. And now

01:18:30.660 you're watching one of your peers pull out what's probably 15 to 20 grams of croutons. Like it's

01:18:37.520 virtually, like it's something that wouldn't even alter your blood glucose. And yet she's saying to you,

01:18:43.920 if I eat these things now, I'm really going to be chasing this for a while.

01:18:49.740 And she said, and I won't be able to figure out the right dose of insulin because of all the fat

01:18:54.120 that's in the salad dressing. She goes, I've just found that it's easier for me to not eat this

01:18:58.700 stuff. So I looked at that and I realized like, this is the ultimate hack. This is somebody who

01:19:03.400 figured out like the ultimate biohack. She's trying to make her blood sugars near normal all the time. 1.00

01:19:08.580 She discovered a group of foods that she needs to avoid. And so she avoids them. 0.68

01:19:12.440 And what did she figure out? I mean, presumably most of what she needed to avoid was carbohydrates,

01:19:17.140 but were there some that she found? And I assume she did not have a CGM at this point in time.

01:19:22.120 That's right.

01:19:22.800 So this is all on finger sticks and this is doing it the really hard way.

01:19:26.780 She's testing 10 times a day or something.

01:19:28.740 Yeah. God, I can't, what did her, do they even have feeling left in their fingers at this point?

01:19:33.880 Quite often they won't. So they go after, if people use finger stick testing, we'll use one spot over

01:19:39.540 and over again and it becomes numb. And that's a way to be able to do it without pain.

01:19:43.540 And I just know from when I'm doing a lot of finger sticking, like the bruising and the

01:19:47.800 discoloration and the discomfort. And, and I think to myself, you know, it's one thing when I'm doing

01:19:53.200 this as a grownup, who's choosing to do this, it's quite another thing for a kid who never signed up

01:19:58.700 for this to be going through this amount of pain.

01:20:02.420 Yeah. And for their parents and for the parents to think about all the other things that could

01:20:06.720 happen. Like for instance, we, you know, unfortunately we have kids who disappear. So

01:20:11.520 we have teenagers who go off to college, you know, and they party with their friends and

01:20:16.760 potentially take the same dose of insulin that they would have taken. And they have fatal

01:20:20.860 hypoglycemic episodes. And that doesn't happen often, but when it does, it's devastating. It's really

01:20:26.740 awful. And that sits in the back of parents' minds. And that kind of anxiety means that nobody ever

01:20:33.940 really just gets to be a parent or a kid because there's this constant fear of what's my blood

01:20:40.820 sugar? Is it high or low? And, and I think of it as a cognitive load. It's the burden of having to

01:20:47.020 think nonstop about your diabetes. And if you can imagine that volatility, the rollercoaster blood

01:20:52.260 sugars that are going up and down, you're going to have to think about it nonstop. And I think that

01:20:58.560 the cognitive load in some ways is probably a really pivotal factor in the anxiety and depression

01:21:05.980 that's associated with type one diabetes. So there's a whole category of a very complicated

01:21:11.260 mental illness that people with type one diabetes have approximately 45 or 50% will have depression or

01:21:18.500 anxiety. And it can be really debilitating. And of course we don't really have any way of knowing

01:21:23.940 is what percentage or what amount or proportion of that debilitating mental illness is resulting from

01:21:31.740 the physiology of the hyperglycemia, the glucose fluctuations and the hyperinsulinemia or the relative

01:21:38.440 hyperinsulinemia versus this much more difficult to track metric of the cortisol levels that accompany

01:21:46.620 the stress or other things that we couldn't even measure if we wanted to, but a company that,

01:21:51.060 as you put it, I like that phrase, cognitive load. I mean, what's the current thinking on that? I

01:21:55.480 didn't realize that the, did you say 40% depression? That seems almost impossible to believe.

01:22:02.880 Well, imagine, let's imagine you have type one diabetes and you've lived with it for 20 or 25 years.

01:22:09.020 And, and right now you're a little bit hungry and you're trying to figure out what your blood

01:22:13.680 sugar is going to be. And you look down at your CGM and you get a number and you're thinking, well,

01:22:17.500 like, what do I have to do? And when do I have to time the dose of insulin and will I get it right?

01:22:22.580 And think about the last time I gave insulin and I didn't give it right. And I had that

01:22:27.600 hypoglycemic episode and, you know, it was embarrassing. And so, I mean, it's a big,

01:22:33.540 really serious challenge in that I try to quantify the, what I call the cognitive load

01:22:38.880 using a diabetes distress scale. And it's an adaptation of a published distress scale. And I

01:22:44.740 basically say to people, okay, I want to talk to you using a Likert scale, a one to 10 scale about

01:22:51.460 your diabetes related distress. And so it's on a one to 10 scale. And one means, you know, you have

01:22:58.440 diabetes, but it's not holding you back at all. It's there. It's in the back of your mind. It's not

01:23:03.420 bothering you. You get to do what you want to do. And a 10 means you have diabetes and you can't think

01:23:08.780 about anything else. And it's pretty much omnipresent and it alters your ability to carry

01:23:14.200 out the things you want to do. So if this is on a one to 10 scale. So what's the median score for a

01:23:21.600 teenager who has had this disease for five years or more? Who's on a high carb diet? Yeah. Seven or

01:23:27.180 eight. And I've seen people who have hemoglobin A1Cs that are approaching 14 who are doing basically

01:23:34.100 nothing to care for their diabetes. Teenagers who are really struggling. And when I describe this to

01:23:40.260 them, their answer is tears pour out of their eyes. And because no one has ever actually asked

01:23:47.560 them, how do they feel? And they will say a 10, but they can't even admit it because no one has ever

01:23:55.440 considered that they have this terrible burden. What they think is, you know, this kid doesn't do what

01:24:01.940 everybody wants him to do. You know, he has got a hard head. He's not listening to us. But in fact,

01:24:07.240 the kid who's not doing anything is thinking about it nonstop and is wracked with guilt and shame. And

01:24:13.220 this is not just true of teenagers. It's true of adults because the adults want to do a better job,

01:24:18.560 but they lack the tools and the support to be able to carry it out and they blame themselves.

01:24:24.280 So this endless cycle of blame and shame and negativity is unfortunately in many cases reinforced by the

01:24:30.900 medical establishment. Because now you come in, you get a hemoglobin A1C of 14. And what do you get?

01:24:35.320 He gets spanked.

01:24:36.280 Yeah, you get a lecture about how you need to do a better job. You're told you're not compliant.

01:24:42.020 You might even be threatened with being fired by your doctor unless you do a better job.

01:24:47.000 You know, you might not be able to make appointments here. This is a privilege to come

01:24:51.460 and see the endocrinologist.

01:24:52.640 And so people end up in this weird cycle of antagonism towards the medical establishment

01:24:59.780 who apparently don't know that much about type 1 diabetes and certainly aren't empathetic about it.

01:25:05.260 And then their loved ones were like, well, why can't he or she do a better job failing to

01:25:10.860 understand how really complicated it is? And there's no well-worn path in the general populace

01:25:16.740 on how to get to a better spot. And so one of the really crucial elements of this, which is so

01:25:24.020 bizarre, is that this diabetes distress or this anxiety or whatever it is, is egocentronic. And so

01:25:32.520 it doesn't feel foreign. People are depressed or anxious, and they just think that this is the way

01:25:38.420 life is going to be forever. They can't imagine another world where they would feel better. They just

01:25:44.000 think that life sucks. And when you get them, I've worked with people, I've actually had the privilege

01:25:50.980 of working with a few people who have dropped their hemoglobin A1Cs in half. And the experience

01:25:57.940 changed my life. You can get somebody from an A1C of greater than 14 down to under 7. And when they

01:26:04.980 talk about it, they're astounded by the changes in their ability to think and feel and, you know,

01:26:10.900 see the world around them and the love that all these people who were trying to help them have

01:26:17.300 that they couldn't even recognize. Yeah, that's very powerful. I don't have anything that I could

01:26:21.960 even think to add to that. Let's go back to the pizza store, pizza shop. You're watching your

01:26:27.380 colleague, what was her name, Mary? It is Marianne Quinn. You're watching Marianne pull the croutons out

01:26:32.280 of the Caesar salad and it plants a bug in your mind, right? Right. So how long from then until

01:26:38.720 you've discovered this continuous glucose monitor little contraption? So only about four or five

01:26:45.080 years. I was a fellow at Boston Children's and I worked in a basic science research lab studying

01:26:49.840 insulin signaling with Morris White. And then I go off and I get my first job as an assistant professor

01:26:54.780 at UPenn and I'm there and I'm, I set up. So, okay. So you're a chop actually, right? So you're 0.98

01:27:01.100 bouncing around between the biggest children's hospitals in the world. And I set up a lab studying

01:27:07.020 beta cells. And my idea was that I was going to use basic science to advance outcomes for people

01:27:12.720 with type one. And I have this background and a bunch of passion in trying to figure out ways to

01:27:17.480 grow new beta cells. And the hope is we'll be able to find some way to, to, to do that safely and

01:27:22.420 reliably. And I also have a clinical practice that's focused on type one and other endocrine

01:27:28.240 disorders. And along the way, I become very close friends with people who live with type one and I

01:27:33.740 start to talk to them and follow them. And I start to meet some of the reps that is the sales

01:27:40.600 representatives for various companies, including one who herself has type one diabetes. This is Natalie

01:27:47.380 Bellini. So she's, she's an amazing person who has type one and she's wearing this device, which

01:27:53.580 is this black hockey puck. It's this Dexcom thingy. What generation was that? It was the very first. 1.00

01:28:01.300 Wow. And so this is approximately 2000, maybe seven. And I thought, well, this is amazing because it's,

01:28:10.980 it's obviously a consumer device it's approved. And then you can give it to patients and they could wear

01:28:16.320 this thing and then they could figure out ways to, to improve their glucose excursions. And actually

01:28:21.920 I, several of my friends went on them and what they discovered was the devices were incredibly

01:28:27.600 frustrating because if you just wear the device and again, a Dexcom or is just one, that's, that's a

01:28:35.780 company that's making a continuous glucose monitor. And what they do is they take a probe that goes into

01:28:40.920 your interstitial space, say in your belly, it's a very small needle and it sits in the space. And

01:28:46.840 what it's doing is it's sampling the interstitial fluid, which is the clear fluid that exists in

01:28:52.120 between cells, which roughly correlates to blood glucose, but is delayed by about eight or 10 minutes.

01:28:57.520 And so the device essentially has glucose oxidase, which is an enzyme and it's stuck onto this platinum

01:29:05.980 wire. And it's basically an enzyme on a stick. And so when glucose comes by and it does the

01:29:11.980 enzymatic conversion, it generates a charge, which is transmitted down the wire and the change in the

01:29:17.780 wire charges is transmitted to this hockey puck like device that tells you your glucose. So that's

01:29:22.640 pretty cool. And it blows my mind when I see it working regularly for patients. And my thought is,

01:29:27.520 okay, so we're going to, we're going to prescribe these for all of our patients and then they're going

01:29:30.680 to figure out how to make their blood glucose is better. But as it turns out, the first generation 0.99

01:29:35.300 of CGMs was a dismal failure and the uptake was incredibly, incredibly small in the population

01:29:42.140 of people with type 1 diabetes. And part of the issue is we didn't really know what to do. So

01:29:46.280 when we looked in the toolbox of trying to help people so they could live with type 1 diabetes,

01:29:51.100 we didn't find much other than badgering them to do a better job. You know, oh, you need to check

01:29:56.780 your blood glucose 20 minutes before the meal and give yourself insulin 20 minutes before the meal

01:30:03.000 meal and do that in a very reliable way so that your blood glucose is normal. But that's not

01:30:08.160 realistic. And for that matter, no one really knows that they're going to eat. So there's a major

01:30:12.760 problem in that we live in this fantasy world in medicine where we think people just need to carry

01:30:18.220 out their lives with more precision and that will be the answer. But the reality is variance is contained

01:30:23.480 in human behavior. It's just what we do. So that idea of badgering people doesn't work.

01:30:28.220 And then maybe it's exercise. So okay, as it turns out, if you're a marathon runner and you have type

01:30:34.340 1 diabetes, your diabetes control is much, much better. And part of the mechanism is via this

01:30:40.500 insulin-independent glucose uptake into skeletal muscle via AMP kinase. So there's two different

01:30:46.920 pathways whereby glucose gets into skeletal muscle. One is via the insulin-dependent pathway. We spoke about

01:30:52.140 that before that goes to GLUT4. The insulin-independent pathway is working through exercise and work of

01:31:00.140 muscle. And AMP kinase is this amazing enzyme that is sensing the energy status of the cell.

01:31:06.580 And if you alter the energy status of the cell through a tremendous amount of work,

01:31:11.060 you can basically drive a bunch of signals that go to tell the GLUT4 to translocate to the membrane

01:31:18.620 in the skeletal muscle. So now you get insulin-independent glucose uptake in the skeletal muscle.

01:31:23.240 So it's still through GLUT4, but it comes from inside the cell activation, I assume, of AMPK.

01:31:30.020 So activating AMPK sends the GLUT4 transporter up. And in a non-athlete, what is the relative

01:31:37.300 difference between insulin-dependent and the AMPK-driven glucose disposal in the muscle?

01:31:43.340 I'm not exactly sure, but I can say that if, for instance, I take a kid who has type 1 diabetes,

01:31:51.320 and we have stable insulin doses, and then they go on like a ski vacation with their parents. And

01:31:58.280 the kids are, you know, mostly playing video games and hanging out. And then they go off to Vail,

01:32:02.860 and they ski for six hours a day. Those kids can require nonstop carbohydrates until their parents

01:32:10.340 figure out that they have a problem. And then they end up on insulin doses that are reduced by two

01:32:14.840 thirds. So the effect can be dramatic. In normal non-exercise associated glucose homeostasis, I

01:32:21.760 think a lot of it is insulin-driven. And so as you carry out more and more exercise, you invoke this

01:32:27.600 pathway more and more. Yeah, that's so interesting. I need to look into this more because I've never,

01:32:32.640 I can't believe it didn't occur to me to look at the patients with type 1 diabetes to answer this

01:32:36.780 question. I've always wondered what the relative difference is in, you know, like if you're taking

01:32:41.800 metformin, which is activating AMPK, and if you're exercising like crazy, you should be able to dispose

01:32:48.040 of glucose with lower and lower insulin levels. Right. And so we can obviously measure this.

01:32:52.540 That's right. In a oral glucose tolerance test. The problem is we don't know why. We don't know what

01:32:58.660 it is that's, is that, that they're more sensitive to insulin, which is, that's a two-hour

01:33:03.820 discussion on end of itself, which you and I'll have over dinner right after this. Or is it that,

01:33:08.700 no, they're, each unit of, or heterodimer of insulin produces the same magnitude of glute

01:33:13.980 for translocation, but they're experiencing a benefit of this AMPK version. In other words,

01:33:21.620 how much of the insulin sensitivity is really less about insulin and more about AMPK?

01:33:26.960 I think for people who are very athletic and who are focused on trying to reduce their body fat,

01:33:32.280 it can probably be a huge contribution. And we have these adults who have type one diabetes who

01:33:36.740 are down to 20 units of insulin and a normal 70 kilogram person would be on 60 or 70 units of

01:33:42.920 insulin. So the marathon runners who are thin, of course they have very thin body habitus, but

01:33:48.980 many of those people are putting in 60, 70, 80 miles a week or on tiny doses of insulin.

01:33:54.660 So again, so we got, I have a patient by the way, with type one diabetes who uses five units of

01:34:00.060 insulin a day. Wow. He walks six miles a day, minimum, sometimes 10, and he does not really

01:34:07.920 consume carbohydrates outside of vegetables. And how much body fat? I've never seen a higher

01:34:13.600 adiponectin and lower leptin in a human I've ever measured. He is a lean, mean machine.

01:34:20.860 Yeah. So those people are, are just my heroes. It's so, I'm thrilled to hear about this.

01:34:24.960 So if we're looking in the toolbox, we find like badgering people to do a better job,

01:34:29.940 maybe exercise. And then like, what else do you got? And so I'm looking around and I'm looking at

01:34:35.440 this and I'm trying to- So going back to the exercise thing for a moment, do you see a difference

01:34:38.720 in, I don't like the term aerobic, but people know what I mean by aerobic exercise versus strength

01:34:43.940 training? Yeah. I think that people who are doing distance, who are, that they actually have

01:34:51.640 more contribution of this AMPK pathway. And Lori Goodyear has studied this for years and years.

01:34:58.340 She's at the Joslin. And again, this, this AMPK exercise associated pathway is incredibly potent,

01:35:05.020 but it's all distance. I don't think it's strength.

01:35:07.960 And then do you think there's a difference between going out and running 10 quarter mile

01:35:12.220 repeats at staggering intensity where you're burning through glycogen in the muscle versus,

01:35:18.900 you know, so that would only be what a total of two and a half miles of running versus going out

01:35:22.820 and running six or seven miles at a much lower intensity. Do you think there's a difference

01:35:26.260 between those two? Yeah. I think there's a, there's a huge difference, but what I hear over and over

01:35:30.020 again is the people who are the, the distance athletes are the people who are able to dramatically

01:35:34.620 reduce their total amount of insulin. And, but again, there's, there's relative debate and this

01:35:40.260 is not well studied. So you'd like to really care to proper study with high intensity exercise versus

01:35:45.500 very slow, long duration exercise. But this idea of trying to sort of hack the system to reduce the

01:35:53.620 total exposure to insulin and therefore reduce the volatility and the hyperglycemic excursions,

01:36:00.460 that's really exciting. And I suspect that the people like your patient who have, you know,

01:36:06.700 down to eight units of insulin, these people have really cracked the code.

01:36:11.220 So when did the Dexcom for you, was it around Gen 4 that it really started to become a real impact?

01:36:18.440 Yeah. The G4 was a special device.

01:36:20.360 So for me, it wasn't just that, that we replaced the hockey puck, at least in my life, that the

01:36:25.900 transformative thing was learning about Dr. Richard Bernstein in his book. And-

01:36:30.960 Do you know Richard well?

01:36:31.980 I only talked to him on a teleconference. I, I'm on a low carb physician teleconference

01:36:37.060 that happens once a month. And I, I talked to him regularly then.

01:36:40.140 You got to get me on this. I, how am I not on this? I'd be very curious to just be a fly on the

01:36:44.400 wall. Will I, is there any chance I can get invited?

01:36:46.900 Yeah. So it's run by Virta Health and it's focused on type one.

01:36:49.980 It's on Virta.

01:36:50.740 Yeah.

01:36:51.380 How did I not know about this?

01:36:53.020 So it's low carb and type one and it's physicians who are very much interested in the intersection

01:36:59.420 of type one and low carb. It's really cool.

01:37:02.680 You know, I've never met Dr. Bernstein, but I've, I've read his book and I've spoken on

01:37:05.600 the telephone with him. He's very graciously extended an offer for me to come out to his

01:37:09.780 clinic and spend time. And I would give anything to be able to make that time. I just haven't

01:37:13.740 been able to, but it's safe to say he's kind of the guy who's spent more time doing this

01:37:19.000 and has more clinical experience in this than probably anyone else.

01:37:23.020 And because he did it with himself, right? It's amazing. He actually got a YSI blood 0.83

01:37:28.820 glucose tester and has been using it at home forever. So he's been blood glucose testing

01:37:36.760 before anybody else did. And he had this essential core idea, which was carbohydrates caused blood

01:37:44.380 glucose excursions. And that he has the so-called rule of small numbers, which says that if you

01:37:49.780 consume more carbs, you need more insulin and therefore there's a greater opportunity to make

01:37:54.740 a mistake. And so if you consume fewer carbs, you will consume less insulin and you'll have fewer

01:38:00.040 mistakes. It's a very simple idea. It's laid out along with a ton of practical tips in his book.

01:38:06.900 He's incredible. He figured this out.

01:38:08.680 And we'll link to his book, but it's called the Dr. Bernstein Diabetes Solution.

01:38:14.320 Yeah, it's a Bible. It's like it's a tome.

01:38:16.480 And it's for type 1 and type 2. And it includes all sorts of pearls, not just on straight diabetes,

01:38:21.700 but also diabetes complications. It's really shocking. And again, this is like the 11th or

01:38:27.260 12th edition. He's been doing this for years and years. So he has a private practice. He's in his 80s.

01:38:32.640 He lives in Westchester County, New York, and is working out and incredibly healthy.

01:38:39.000 So he figured out that if you avoid carbohydrates and if you give medium and long-acting insulin to

01:38:44.620 cover protein and fat and basal metabolic requirements, that you could get blood glucoses

01:38:50.720 very, very close to normal. And he typically prescribes a low-carb, high-protein diet.

01:38:57.060 It's not low-carb, high-fat. It's not ketogenic.

01:38:59.520 What have we learned from Bernstein's experience on how insulogenic protein is? Because that's

01:39:04.820 somewhat debatable, right? I mean, clearly, protein stimulates some insulin. And certain

01:39:10.140 in vitro studies suggest it can be very insulogenic. What does the real-world experience

01:39:16.900 tell us about that?

01:39:18.040 In my patients with type 1 diabetes who consume protein and don't cover it with insulin,

01:39:22.520 they can get these massive glucose excursions. And we think in general, there's a ratio of around

01:39:27.140 10 grams of protein to end up being 6 grams of carbohydrate. But the problem is the kinetics

01:39:34.900 are not immediate. So you don't get gluconeogenesis in a matter of seconds. It's delayed over hours.

01:39:41.280 If we had a-

01:39:42.680 And doesn't it also depend on their nitrogen requirement? So in other words, you could have,

01:39:47.060 which is obviously true with glucose, but you could have two individuals who are similar

01:39:52.200 at the crude metrics. Say each of them gets 40 grams of protein, but one of them has just finished

01:39:59.440 exercising or lifting weights and the other one hasn't. And you're going to presumably see very

01:40:03.640 different glucose responses, correct?

01:40:05.620 But for the vast majority of humans, we're at homeostasis. And overall, if you look at the USDA

01:40:11.200 requirements, this is a suggestion we need to consume 56 grams of protein. And the net result

01:40:18.140 of that is approximately 6 grams of protein that go into metabolism. So, excuse me, 7 grams. So

01:40:25.900 there's a net excess that's 49 grams that ends up being approximately, what is that, 30 grams of

01:40:31.800 carbohydrate. But on the other hand, if you're consuming high protein, if you sit down to a 12

01:40:37.180 ounce, you know, ribeye, and it also has fat, which will delay the absorption, the sum total of that

01:40:43.440 could be 80, 100 grams of protein that turns into glucose, that is at this massive slow wave that

01:40:51.180 rises and stays high for much of the evening after you eat it, which could be very frustrating. So a lot

01:40:57.820 of people who with type 1 diabetes think, well, I'm going to do this sort of low carb thing. So

01:41:02.560 instead of eating the pasta, I'm going to have a piece of fish and some vegetables, but there's no

01:41:10.140 carbohydrate. Cool. I don't need to take any insulin. But then you get this weird glucose excursion that

01:41:16.040 comes from the protein. And the essence of the Bernstein method, if you read his book, is to try

01:41:21.160 to cover protein with insulin. And he uses regular insulin, not short-acting hemolog or novolog. And the

01:41:28.980 idea is the regular insulin, which is human insulin, has a peak of onset within approximately

01:41:35.260 an hour or two, and it lasts for six to eight hours. And so it roughly, very roughly corresponds

01:41:43.760 with the kinetics of what protein will do to your glucose. So what's unique about Bernstein is he's

01:41:48.880 using this ancient form of insulin, human, regular human insulin, instead of these very rapid acting

01:41:54.660 analogs, because the protein is absorbed and turns into glucose in a delayed fashion.

01:42:00.560 I've never really understood these rapid onset insulins because they seem to get back to this

01:42:05.780 oscillation problem of, I mean, so the first observation of Bernstein is more carbohydrate

01:42:11.680 means more insulin, more insulin means more vacillation. But now you're introducing another

01:42:15.980 effect, which is short-acting insulin causes more rapid changes in the glucose, which can more

01:42:23.700 easily result in you making another decision after that amplifies the signal instead of dulls the

01:42:30.020 signal.

01:42:30.320 Right. So to go back to your analogy of the plane, potentially what it means is a plane that can

01:42:35.240 much more rapidly gain or lose. Yeah. And you can get into a lot of trouble. And unfortunately,

01:42:42.140 the insulin pump companies and even the insulin companies have advocated an approach that says,

01:42:47.060 look, people who live with type 1 diabetes have tough lives. We need to support them and they should

01:42:52.580 be allowed to eat what they want and they should just cover for it. And the problem is you end up with

01:42:57.220 these crazy oscillations. And the oscillations are associated again with a tremendous burden of

01:43:01.840 illness and this cognitive load and this fear and this anxiety.

01:43:04.560 And you'll have to forgive my sort of snarky-ism, but that's a really wonderful business model if you

01:43:11.100 want to sell more insulin too.

01:43:12.360 Yeah, exactly. Because you consume, well, first of all, everybody gets a pump and then they're told that

01:43:17.500 this is the way to feel normal. You're on shots, but you go to pump and then you can just eat what you

01:43:22.080 want. And then, and then, then of course you consume insulin and you gain weight and you need

01:43:26.860 more insulin. So, yeah. And I'm not, I am so not the conspiracy guy, like, but I also believe it's

01:43:34.680 important to understand incentives. And I don't think pharma is evil. I think far more good has

01:43:40.580 come from the ability of being able to sell insulin and sell insulin pumps and all these things. I really

01:43:45.360 do think these are a net positive, but I also think we have to keep in mind there, there are modifications

01:43:51.680 that you can make. And obviously you live this with your patients that can give you 80% maybe of the,

01:43:59.120 of the joy that comes from eating. I mean, let's be honest. I think to eat a restricted diet, to eat a diet

01:44:04.200 that is far fewer in carbohydrates, you're giving up some pleasure. There's just no doubt about it. But if you

01:44:10.080 could give up 20% of the pleasure of eating and get 80% more benefit in your health, you know, those

01:44:15.360 are the kind of asymmetric benefits that I think in the long run probably, you know, pay off, but,

01:44:21.320 but it's still not easy. I had to figure this out for myself and I had seen people do low carb and I

01:44:26.280 didn't really understand it. And I just started doing it myself. So I gradually cut back carbohydrates

01:44:32.420 until approximately four years ago when we went low carb as a family and my wife cleared out the pantry.

01:44:39.400 There were no more enriched carbohydrates in our house, but we had teenage girls who were, 1.00

01:44:43.840 one was in college and one was going off to college. And like, I can, I can remember the day

01:44:48.960 and I had wanted to do it. She wanted to do it. We just decided to try it. And so we went low carb

01:44:55.280 as a family and then we've stayed on it ever since. So actually these days I'm probably ketotic.

01:45:00.940 I wouldn't be surprised if my beta hydroxybutyrate is well above one right now.

01:45:04.600 Yeah. So when did you start to, I mean, I guess now that I'm thinking about it, it was,

01:45:10.280 we were probably at the G5 by the time you and I met. And I remember you showing me data. And I

01:45:15.640 really think at the time your patients averaged about 5.7 in their hemoglobin A1c, which is as

01:45:21.380 normal as normal gets.

01:45:22.620 Yeah. And part of it, you know, it took me a while to understand the power of this.

01:45:27.040 And you also noted by the way, that they had fewer hypoglycemic events

01:45:30.500 than patients who were walking around at seven.

01:45:32.860 Yeah. The incredible thing is that you can reduce the excursions and people have very few lows. And

01:45:38.060 again, I read Bernstein and I also learned of a, of a recent Facebook group, which is called type one

01:45:44.960 grit, T-Y-P-E-O-N-E-G-R-I-T. And so that's a group of people who are followers of the Bernstein

01:45:52.460 method. And it's an amazing group of people who are supporting each other. And you see CGM tracings

01:45:57.500 and you see foods that they eat and you see happy kids or adults, and you might see somebody

01:46:04.180 complaining about their healthcare provider who didn't have a clue about low carb. And it's just

01:46:09.040 sort of a conversation that happens in private amongst people who live with type one or those

01:46:13.960 who are, who have a loved one. And there's also healthcare providers like me who are part of it.

01:46:19.420 And again, so it took me a while to understand it and to understand its nuance and to start to

01:46:24.240 iterate it. But for me, the real turning point was changing the way I practiced medicine. So

01:46:30.920 the problem with endocrinology is the schedules were so tight. And the only way I could see patients

01:46:37.220 was every, you know, 20 minutes or 30 minutes. And to try to talk to somebody about something as

01:46:42.680 complicated as living with type one, three or four times a year, and also renew their prescriptions and

01:46:48.520 go through their labs and document and all this stuff was impossible. And I just decided for my

01:46:54.640 own sanity that I needed to change my practice. And so at the time I had a very small amount of

01:47:01.140 clinical responsibility because I was mostly in the research lab and I also had administrative

01:47:04.900 responsibility. So what I decided to do was to make sure that when I saw patients, I only saw them

01:47:11.300 at the end of the day. And the idea was I'd see them starting at 430 and we would just spend as much

01:47:18.280 time as it took. And then I would, I added on other patients. But for instance, you know, some

01:47:24.780 endocrinologists will run from room to room to room to room. I only need one room because I'm going to see

01:47:30.880 a patient for an hour. And then when we're done, they leave and the new one comes in. I don't need to run

01:47:36.360 back and forth. There's nobody else. So I made a conscious decision that I was going to be more

01:47:41.240 inefficient, that I was going to spend more time with people. It's possible I was going to spend much

01:47:45.920 more time doing clinical work for the few patients that I saw. But I wanted to carry out experiments

01:47:51.540 just to try to determine what sort of impact I could have. And the crux of the idea was if you have

01:47:58.680 limitless resources that you make available to a patient and you really try to listen and understand

01:48:06.080 and support people, could you have transformative outcomes? And fast forward several years, what I've

01:48:12.680 discovered is that some people who medicine is essentially given up on, you know, these teenage

01:48:17.780 kids who have hemoglobin A1Cs of 10 or 12 or 14, who are basically on a superhighway to death. You know,

01:48:25.080 these are the people who are going to have their first heart attacks when they're 35.

01:48:27.820 that you can do a lot and help them and they feel paralyzed and they're looking for somebody who

01:48:35.560 can really understand how difficult it is and to provide a level of comfort and compassion. And

01:48:42.200 sometimes the first five visits, you end up talking about almost nothing at all.

01:48:46.620 But you're establishing a rapport and they get that you're going to be in a different mold than maybe

01:48:50.880 the other endocrinologists they've seen.

01:48:52.340 And so, you know, I try my hardest and this is really hard for me because I love to talk,

01:48:56.900 but you know, I try my hardest not to talk and to just ask open-ended questions and to be there.

01:49:02.020 And in general, I found that if I'm asking questions and allowing them to have the space

01:49:08.980 that we can open up and ultimately they can begin to acknowledge that it's really difficult.

01:49:15.860 They're looking for new ways. They're eager to try it, but they are distrustful of this idea that

01:49:21.320 I'm just going to sort of give them a short list like a recipe and then run away and expect them

01:49:26.480 to do it. So it requires patience. And this is especially true of teenagers who, you know,

01:49:31.880 that feedback loop of giving them information and seeing them act on it is delayed by all their

01:49:37.600 growth and development and everything else. And especially with teenagers, it's so important

01:49:42.240 to suspend disbelief and to be there for them in a patient and loving way, even if they can't

01:49:49.300 immediately respond in the way you want. So we now have in the toolkit of treating patients

01:49:54.760 with type one diabetes, we've got exercise, we've got continuous glucose monitoring, we've got

01:50:02.020 carbohydrate restriction. Are there any other apps or technologies or things that you think are

01:50:09.400 paramount to helping or foods for that matter, meaning packaged foods that are particularly useful?

01:50:15.500 You know, in other words, if you're a parent and you're listening to this, or if you're a person

01:50:19.360 and you're listening to this who has type one diabetes and you just, this is all news to you,

01:50:23.040 you're listening to this and you're, this is totally blowing your mind. What do you need to go out and

01:50:27.280 get to get started?

01:50:28.700 Well, I mean, CGM is essential.

01:50:31.060 And I'm assuming, are there any insurance companies that are not covering CGM for type one?

01:50:35.880 Plenty.

01:50:36.540 Jesus Christ. Seriously?

01:50:38.220 Yeah. Well, part of it is a huge fraction of people in our country now have these high deductible plans.

01:50:43.380 So they're on the hook for their healthcare expenses. And when their kid is healthy and

01:50:48.080 everyone's healthy, they're good. And all of a sudden the kid gets diagnosed with type one

01:50:50.840 diabetes and it's like $10,000 a year. Yeah. $10,000 a year.

01:50:55.320 Right. The insulin cost alone can just kill these families. 0.99

01:50:58.120 And the CGM. So, but CGM is really important. And then in the toolbox, there's a few other key

01:51:04.600 things, but the most important thing is information. So you've got to have Bernstein's book.

01:51:09.080 You should be following type one grit, which is this Facebook group, which again,

01:51:14.000 which is a wonderful place to learn about low carb. There are wonderful podcasts and also some

01:51:20.540 terrific YouTube videos talking about the intersection between low carb and type one.

01:51:25.640 I want to put in a plug for an organization called low carb down under, which is really a

01:51:31.420 wonderful group of people. Dr. Rod Taylor in Australia, in Australia. Yeah. Dr. Rod Taylor is an

01:51:36.740 amazing character. He's an anesthesiologist in Melbourne and a passionate advocate and supporter

01:51:42.880 of all things low carb. And his organization goes around the world doing AV capture of low carb talks.

01:51:51.880 And so on YouTube, they have a YouTube channel with over 8 million views on various low carb things,

01:51:59.100 including some terrific talks on low carb by, for instance, Dr. Troy Stapleton, who's a radiologist

01:52:04.640 who got type one diabetes himself and figured out how to hack it using low carb. Where does he live?

01:52:10.200 He is in Brisbane. Okay. He's an amazing person. Very nice. So there's this growing community of

01:52:16.640 low carb nerds who are trying to help people with type one. And again, there's a bunch of different

01:52:21.780 places where it's happening. Another really nice resource is Adam Brown's book, which is called

01:52:27.540 Bright Spots and Landmines. And he works for Close Concerns. And he wrote this book essentially as the

01:52:34.020 manual for diabetes that he wished he had been given. And it contains lots of tips and tricks about how to

01:52:41.420 think about living with type one. And it's from wellness and to exercise, to sleep, all the way

01:52:49.180 to very practical tips around low carb and nutrition and type one.

01:52:53.520 How careful does someone with type one diabetes need to be if they go on a ketogenic diet?

01:52:58.620 Obviously, in patients with a normal pancreas, even those who are somewhat compromised by type

01:53:04.480 two diabetes, I'm not aware of any case of ketoacidosis, diabetic ketoacidosis. But in

01:53:10.480 patients with type one who have no insulin production, that would have to be a real concern,

01:53:14.280 wouldn't it? Well, I think that there are some interesting intersections in between a hardcore,

01:53:18.980 I'm not talking about generic low carb or low carb high protein, all of the Bernstein method.

01:53:24.380 But let's talk about the people who are trying to get into nutritional ketosis with a beta

01:53:29.740 hydroxybutyrate that hovers around one to two. Well, those people have a couple of unique problems. And

01:53:35.600 one is diabetic ketoacidosis, as you mentioned.

01:53:39.080 Again, for the folks to understand why that's the case, and you or me, even when I'm fasting,

01:53:44.300 and I fast for a week at a time, my beta hydroxybutyrate level will reach six or seven

01:53:49.200 millimolar. But it's not going to reach 10 millimolar. It's certainly not going to reach 12

01:53:53.880 or 14, which is where we start to see the acidotic changes. Because eventually, that ketone is going

01:54:00.260 to elicit some response from insulin. In the person who has no beta cell reserve, that ketone

01:54:08.100 can go unabated. So prolonged fasting in a person who lives with type one diabetes is probably a bad

01:54:13.700 idea. Because you could get your beta hydroxybutyrate to six or eight or 10, and you could

01:54:17.560 start to feel sick. Another issue is, it's a new phenomenon called euglycemic DKA, euglycemic

01:54:26.260 diabetic ketoacidosis. And we're seeing this in these sodium glucose co-transporter inhibitors,

01:54:33.760 these SGLT2 inhibitors. And it's a very bizarre phenomenon. But in people with type one diabetes who go

01:54:40.320 on these sodium glucose co-transporter inhibitors, they can get a strange form of DKA where their

01:54:47.940 glucose is not elevated. And it has confounded the field. And there have even been deaths

01:54:54.380 associated with this. There's currently clinical trials. And full disclosure, I've been a consultant

01:54:58.780 on one of these programs.

01:55:00.860 Does this only occur in patients with type one diabetes? Or can patients with type two diabetes,

01:55:05.720 who are the predominant users of the SGLT2 inhibitors also get this euglycemic ketoacidosis?

01:55:12.340 We think it's mostly type ones, but it may also be type twos. And the problem, I think it's probably

01:55:18.080 mostly type ones. And here's the issue. Like, how do you get this? Patients with type two can get DKA,

01:55:25.280 but largely it's at onset. It's not established disease. The people who have type one who have

01:55:32.520 decided to take these drugs. By the way, this is all off-label and I'm not at all advocating it.

01:55:36.460 There are clinical trials and we're trying to develop appropriate safety and monitoring for

01:55:41.340 this. Listeners shouldn't be given free license to go on these things because this is really

01:55:45.360 complicated. But again, if you're on a really potent glucose lowering agent, which is what these SGLT2

01:55:52.700 inhibitors are, and you have type one diabetes and your insulin is infusing from your insulin pump

01:55:58.180 into your belly and the catheter plugs and it stops infusing insulin, then normally you'd look

01:56:04.480 at your CGM and you'd see, okay, my glucose is rising. I must have a problem. But if the glucose

01:56:09.640 lowering agent is essentially dragging all that glucose out into your urine, essentially what it

01:56:16.900 does is it's dragging away glucose, which is the primary biomarker of life-threatening insulin

01:56:22.420 deficiency. Elevated blood glucose is the primary biomarker of life-threatening insulin

01:56:27.040 deficiency for people who live with type one. So when you look at your glucose and you see that

01:56:31.940 it's normal and you're on an SGLT2 inhibitor, it creates a situation of what I would call cognitive

01:56:38.300 dissonance, where people just don't understand that they're in trouble. And there may be a delay

01:56:43.100 in between when they had the plug in their catheter or some other interruption of insulin infusion

01:56:49.960 and when they actually end up recognizing it. And so I think that they end up receiving medical

01:56:54.580 care way later. It's also possible that they have elevations in ketones from these SGLT2 inhibitors.

01:57:01.760 The beta-hydroxybutyrate is a little bit higher in the SGLT2 treated population compared to the

01:57:08.700 controls. Okay. So that phenomenon of euglycemic DKA might also extend in rare cases to people with type

01:57:17.340 one diabetes who are on nutritional ketosis. And the way I think of it is some of them may have depleted

01:57:22.560 glycogen because they're not fully fat adapted. And many of them will have very little carbohydrate

01:57:27.740 in their intestines because they just don't consume that much carbohydrate. So those people,

01:57:33.180 if they have a plug in their catheter, might not see the same sharp rise in blood glucoses

01:57:39.640 that you normally would, and they might have a delay.

01:57:42.680 So this is all associated with pump use. In other words, we don't see this in people who are not using

01:57:47.200 a pump. Well, we also see it in people who forget to take their shots. And oddly enough,

01:57:52.080 people with type one diabetes do forget shots on occasion, especially the teenage kids that I've

01:57:56.960 cared for. So God, there's so many more things I want to talk about. Well, one thing I want to just

01:58:01.540 mention is the intersection of nutritional ketosis and hypoglycemic awareness. So there are people who

01:58:07.540 are in hardcore ketosis who have circulating beta-hydroxybutyrate or one or two, who become

01:58:13.540 completely immune to low blood glucoses. And I had a patient call me up and he goes, Dr. Jake,

01:58:20.980 guess what my blood sugar is? I said, what? He goes, 25. I said, why are you calling me and not

01:58:28.320 taking carbohydrate? He goes, I feel fine. So he- What was his BHB? Oh, like two or three.

01:58:34.840 So he had done the George Cahill experiment where he had- Except Cahill subjects were like six

01:58:39.740 millimolar, seven millimolar. Two or three is relatively low to be thriving at 25 milligrams

01:58:46.620 per deciliter. That's barely over one millimolar. And so he has hypoglycemic unawareness. And so to

01:58:52.400 quote Peter Thiel, like, is this a feature or a bug? If it's a feature, well, he's hypoglycemic

01:58:58.540 unaware, but he's fine. He's cruising around and he had the gumption to call me up. If it's a bug,

01:59:03.600 well, maybe the delta in between his blood glucose when he called me up with a blood glucose at 25

01:59:08.540 and a fatal low might be relatively small. I have no idea.

01:59:13.500 There are many other things I want to chat about, but based on some time constraints,

01:59:18.040 I know we need to wrap this up. So maybe we'll have to come back and do a discussion about two

01:59:23.540 other things I wanted to get into, which is sort of the future of this. I certainly remember being

01:59:28.020 involved in the care of patients who got pancreatic transplants. That's obviously a transient

01:59:32.820 solution that often comes accompanied with a kidney transplant. There are certainly beta cell

01:59:37.880 transplants, which have really not been that successful though. It's always one of those

01:59:41.320 things where it seems like it's just a year away from being perfect. There's xenotransplantations 1.00

01:59:47.100 of beta cells when you take the beta cells from another animal or something that's non-human that

01:59:52.500 has all of its own problems associated with the immune response. You talked a little bit about stem

01:59:57.180 cells. And then of course we've alluded to, you know, fully implantable or even non-implantable,

02:00:03.280 but fully autonomous pancreases. So that's a whole other topic that would be great to come back and

02:00:08.240 talk about. But I'd like to close on one, one other question, which is if you were to reflect on

02:00:13.460 everything you've learned in caring for patients with type two diabetes, what are the most type one,

02:00:20.080 sorry, type one, I'm sorry, type one diabetes. What are the most important lessons that would

02:00:25.480 extrapolate to the non-diabetic population?

02:00:28.500 When I look at the cardiovascular outcomes and the weight gain associated with hyperinsulinemia

02:00:33.820 in type one, I'm really shocked. And again, I think you have this unique population where you

02:00:38.920 see the insulin that goes into them and you see its direct impact on weight gain. So we know that people

02:00:45.700 consume very high carbohydrate diets and gain a bunch of weight in type one diabetes. They're at risk

02:00:52.040 for cardiovascular illness. And to me, that just says that there's something about the standard

02:00:56.180 American diet that's making us sick. And ultimately we will be at risk for cardiovascular disease. 0.99

02:01:03.760 That's the primary thing that humans will die from.

02:01:06.240 What do you think it is about hyperinsulinemia that seems to drive this? Because this is the thing

02:01:12.140 that I've always found fascinating is that, so when you look at something like, so let's look at the

02:01:16.860 type two diabetic cohorts. When you have patients who are using drugs like metformin, which are lowering

02:01:22.840 glucose to achieve better outcomes versus drugs that raise insulin, because the patients with type

02:01:31.940 two diabetes still have the ability to make insulin, you can, in both of those examples, get the same

02:01:37.680 glucose level. But in one patient, there's a higher insulin level. Those patients have equal

02:01:42.660 microvascular outcomes, but they have different macrovascular outcomes.

02:01:47.980 Which is a sort of natural experiment to suggest that the insulin is playing a bigger role on the

02:01:54.720 macrovascular disease. The glucose is playing the role in the microvascular disease.

02:01:59.420 Yeah. And in the case of these STLT inhibitors, you know, they're the first drugs that have really

02:02:04.980 shown a profound impact on cardiovascular outcomes. This is secondary prevention for cardiovascular disease

02:02:12.920 in people who have already had an event, who also have type two diabetes. So,

02:02:19.500 empicliflozin is an amazing drug when reduced the cardiovascular complications by 40%. And the drug

02:02:26.860 essentially drags glucose into the urine, and it suggests that circulating insulin values themselves

02:02:32.400 might be in some way contributing to cardiovascular disease. The simple answer is, I don't know,

02:02:38.180 there's a lot, there's a rich debate.

02:02:39.360 Yeah. I mean, do you think it is due to endothelial injury?

02:02:42.840 Some people claim it's all about the kidney and fluid and the amount of salt retention and

02:02:48.760 something else. But to me, as a basic scientist who grew up idolizing Dr. Cynthia Kenyon and the

02:02:55.780 amazing work that she did in C. elegans from the Daft-Dower complex, those papers suggest that when you

02:03:02.320 make a modest alteration in insulin signaling, the worm lives twice as long. And that fundamental

02:03:08.760 observation has now been borne out in model organism after model organism.

02:03:13.740 But she did that, I mean, it's tough to, I mean, I'm so familiar with Cynthia's work,

02:03:18.500 so I don't want to bore people with all of my nuanced thinking on this. But it's hard to make

02:03:24.340 that exact comparison to us, right? Because that was a modulation of DAF2 and DAF16. So they were,

02:03:30.080 she was upping FOXO effectively and downing IGF. And to get the real step function change in longevity,

02:03:38.760 she had to calorie restrict the worms as well. So there are so many moving pieces in those models

02:03:45.880 that I guess, well, if, look, if you don't know the answer, that means it's, we've still got some

02:03:50.260 thinking to do on this.

02:03:51.500 But I want to go back to your patient, the one you told me about who's super healthy,

02:03:55.360 who's on six units of insulin. I just, I think that those people may be telling us something.

02:04:00.920 He's thin, he's active, he's on very little insulin, and he's quite vigorous. And I just

02:04:06.700 wonder whether carrying out these tricks to try to reduce the, your circulating insulin could be

02:04:12.900 incredibly beneficial.

02:04:15.200 Well, part of the reason I love the CGM is it's the best proxy I have for what I don't have,

02:04:20.960 which is an area under the curve of insulin. I'd love to, some people ask me sometimes,

02:04:24.940 if you could wave a magic wand and have something that doesn't exist, what would it be? And

02:04:28.180 my first answer is a metabolic signature for autophagy. My second answer would be every night

02:04:34.260 I'd like to go to bed and have a little readout that says, this is the total amount of insulin

02:04:39.280 your pancreas made in the last 24 hours. Can you imagine adding that to the list of my aura ring,

02:04:45.400 the CGM, all the other things I think are really, really driving benefit in terms of health,

02:04:51.020 but to be able to actually know the AUC, the area under the curve of insulin produced every day

02:04:55.480 would be remarkable. And amazingly, someone with type 1 diabetes gets that for free.

02:05:00.760 They just have to look at the syringe and know what they injected.

02:05:04.120 Yeah.

02:05:04.260 So if there's one thing that that patient with type 1 diabetes has over the rest of us that don't,

02:05:09.000 it's that they can know at the end of the day exactly what their AUC of insulin is,

02:05:13.580 while we can only estimate it tangentially by looking at the things we look at, which are,

02:05:19.200 you know, meaning someone like me can look at average glucose and standard deviation of glucose

02:05:22.980 and just try to minimize those as an indirect proxy for that AUC of insulin.

02:05:28.180 These people are really amazing. And the most successful are so dialed into their physiology.

02:05:33.540 They're doing quantified self before it was even a term.

02:05:36.960 Yeah, exactly. Well, on that note, Jake, I know you're super busy here in San Diego

02:05:43.440 today. And so I really appreciate you making the time to come over here and share a little bit of

02:05:48.460 your wisdom and your insights. We'll have to do this again at some point, because as I said,

02:05:52.100 there's so many of the things that I want to talk about, and I'm sure there are other people

02:05:54.860 listening to this who do as well. So with that said, until next time.

02:05:58.380 And thank you. It's an honor. I really love your show. I've been listening every week.

02:06:01.400 You can find all of this information and more at peteratiamd.com forward slash podcast.

02:06:08.880 There you'll find the show notes, readings and links related to this episode. You can also find

02:06:13.980 my blog at peteratiamd.com. Maybe the simplest thing to do is to sign up for my subjectively

02:06:19.220 non lame once a week email, where I'll update you on what I've been up to the most interesting

02:06:23.700 papers I've read and all things related to longevity, science, performance, sleep, et cetera.

02:06:28.600 On social, you can find me on Twitter, Instagram, and Facebook all with the ID peteratiamd, but

02:06:34.980 usually Twitter is the best way to reach me to share your questions and comments. Now for the

02:06:39.100 obligatory disclaimer, this podcast is for general informational purposes only and does not constitute

02:06:43.740 the practice of medicine, nursing, or other professional healthcare services, including the

02:06:48.640 giving of medical advice. And note, no doctor patient relationship is formed. The use of this

02:06:54.180 information and the materials linked to the podcast is at the user's own risk. The content of this

02:06:59.180 podcast is not intended to be a substitute for professional medical advice, diagnoses, or treatment.

02:07:04.680 Users should not disregard or delay in obtaining medical advice for any medical condition they have

02:07:09.200 and should seek the assistance of their healthcare professionals for any such conditions.

02:07:14.320 Lastly, and perhaps most importantly, I take conflicts of interest very seriously. For all of my

02:07:19.080 disclosures, the companies I invest in and or advise, please visit peteratiamd.com forward slash about.

The Peter Attia Drive - February 18, 2019

#41 - Jake Kushner, M.D.： How to thrive with type 1 diabetes and how everyone can benefit from the valuable insights

Episode Stats

Summary

Transcript