The Peter Attia Drive - #52 - Ethan Weiss, M.D.： A masterclass in cardiovascular disease and growth hormone - two topics that are surprising interrelated

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 can trust me

00:01:59.700 if I'm telling you about something when you know I'm being paid by the company that makes it to tell

00:02:04.700 you about it. Another reason selling ads doesn't feel right to me is because I just know myself. I have

00:02:10.900 a really 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.

00:03:32.800 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

00:03:57.880 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 my good

00:04:13.540 friend, Dr. Ethan Weiss. Ethan and I have known each other for a few years. We both sit on the advisory

00:04:18.240 board of a company called Virta Health, and we've always found ourselves sitting around the table

00:04:23.100 at these meetings or elsewhere, just having super nerdy discussions. And we just decided we got to at

00:04:29.820 least put some of these discussions on the podcast. So what follows is a really deep and for me,

00:04:37.200 certainly interesting discussion about two topics that at the surface seem completely unrelated,

00:04:43.420 cardiovascular disease and the role of growth hormone and IGF in disease. Now it won't be clear

00:04:50.140 from what I just said, why one guy would be an expert on both of those things. But as the story unfolds,

00:04:56.860 you'll see that Ethan is quite a unique individual. By training, he's a cardiologist. He specializes in

00:05:02.280 preventative cardiology at UCSF, trained at Hopkins in both medical school and in his residency, and then

00:05:09.680 completed his fellowship at UCSF where he has since remained. And his interests are wide ranging from

00:05:15.460 prevention and all aspects of it, which includes lipids and the management of blood pressure and the

00:05:21.320 tendocrine system, et cetera. But what I found most interesting in this discussion, in addition to just

00:05:26.180 the master's class in understanding everything from acute coronary syndrome to all of the complex

00:05:33.760 nuances around stent placement, which is something that I think anybody listening to this knows at least

00:05:39.240 somebody who's had a stent placed and to sort of go through all of that literature in detail and

00:05:43.860 understand who are the ideal candidates versus who is not. All of that, I don't want to say paled in

00:05:49.520 comparison because I would minimize it, but it was nothing compared to, at least for me, diving into

00:05:55.480 what led him to pursue this field of endocrinology and specifically what he found with respect to

00:06:04.280 growth hormone and IGF. And this is something I only knew about Ethan maybe six to nine months ago.

00:06:09.260 It was really quite recent that this came onto my radar about his level of expertise in this.

00:06:13.120 And so we just spent a great deal of time talking about this. And you could almost think of this as

00:06:16.980 two separate podcasts, one that really deals with cardiovascular disease, specifically

00:06:22.040 interventional radiology and diagnostic techniques. We talk extensively about calcium scores and CT

00:06:28.280 angiograms and things like that, that we get asked about all the time. And then there's basically a

00:06:32.300 second podcast here, which is a really interesting discussion around growth hormone, its effect on the

00:06:37.280 liver, the brain, et cetera. Finally, we touch really briefly at the end on a company that Ethan has

00:06:43.140 started quite recently called keto or key to depending on how you pronounce it, which is a

00:06:48.140 breath analyzer for acetone. So again, nothing that novel there, but what is novel is the way this

00:06:55.580 device works and socially kind of provides feedback and allows you to get great feedback when you're

00:07:02.680 fasting or on a ketogenic diet or something like that. So I hope you enjoy this podcast. This is one of

00:07:06.980 those ones where the show notes are really going to be helpful, especially for people who actually want to

00:07:11.800 go back and look through all of the literature in all of the clinical trials we talk about,

00:07:16.540 and even just some of the diagrams, frankly, to even understand when we're talking about EKG

00:07:20.680 changes and things like that. Sometimes the diagrams help. So without further delay,

00:07:24.540 please welcome my guest, Dr. Ethan Weiss.

00:07:29.320 Ethan, thanks so much for making time today.

00:07:31.600 Thank you, Peter. Thanks for coming over to San Francisco.

00:07:34.740 I can't believe this is the first time I've been to your office, which I must say,

00:07:37.440 anyone who knows me knows how much I love whiteboards, but the glass whiteboard is just

00:07:41.880 next level stuff. We got a little bit of flack when we picked these out. The architects,

00:07:46.620 this is actually super fancy Italian glass and the architects sort of convinced us you'll really

00:07:52.360 appreciate it. And I love it. Actually, my kids love it too. They've daked out their own territory on

00:07:58.060 this. I actually like that the kids have their drawings adjacent to the drawing of mTOR.

00:08:02.580 Yeah, perfect. I think you should always be able to mix those. The other thing that I don't know,

00:08:07.120 I guess it's because I've never been to your office that I don't know how we have never

00:08:10.680 discussed is our mutual appreciation for Wayne Gretzky. You have in front of me a home jersey

00:08:17.140 circa 1986 Edmonton Oilers signed by none other than the great one.

00:08:25.620 Yeah. Look, I grew up in Baltimore, which was not really a hotbed of hockey. I think they were the

00:08:32.100 skipjack, minor league hockey team, triple A team, I guess. I don't know if they call them that in

00:08:36.660 hockey. Anyway, we had the skipjacks and the clippers and I never really was too into hockey.

00:08:41.000 I was a baseball fan and a football fan, but my dad had grown up in Chicago and really did like

00:08:45.940 hockey. And one year we were back there visiting my grandparents for Christmas. I think it was the

00:08:51.080 winter of 1979. And he said, let's go to a Blackhawks game. So we went to the old Chicago

00:08:54.980 stadium. It was my first introduction to this game that, you know, I basically fell in love with

00:09:01.020 instantly. And at that time in the late seventies, early eighties, it was, you know, the birth of

00:09:05.700 these dynasties, I think first the Islanders and then the Edmonton Oilers. And you couldn't be a

00:09:10.820 sports fan, but you couldn't be a hockey fan and not appreciate Gretzky. So I eventually developed

00:09:14.920 into being a Washington Capitals fan, which was sheer torture until last June, which was super fun.

00:09:20.580 But in the process of being a fan, of course, I always remained true to Gretzky. Anyway, a couple of

00:09:24.800 years ago, my mom said, Hey, look, I'm coming to California. We're going to go to LA. My uncle,

00:09:30.920 my brother, my uncle is an oncologist down there. He's being honored by the sarcoma foundation.

00:09:35.900 I said, sure. It's an hour flight. I can hop on. I'll come down. I'll join you. So we go,

00:09:39.160 it's one of these LA galas. And there's like a big series of tables with, with a bunch of silent

00:09:44.300 auction items. And I walked by this Jersey a couple of times. And on the third or fourth time I walked by

00:09:50.020 the guy who was manning the table stopped me. And he said, Hey, you know, buddy, I see you walked by

00:09:53.900 that a few times. Do you have any interest in it? I said, yeah, I've got interest in it, but there's,

00:09:58.220 I can't, I can't afford that. And he said, nah, listen, I'm pretty sure if you bid the minimum,

00:10:03.080 you'll get it. And the minimum, minimum bid was a hundred dollars. And I said, you gotta be kidding

00:10:08.240 me. Gretzky played here in LA and how can, that's impossible. It's a signed Jersey. It's beautiful.

00:10:14.520 The frame alone is worth more than a hundred dollars. So he finally convinced me to do it. I put down the

00:10:19.080 bid and I walked out with this Jersey, which my wife instantly told me she loved, but she loved it

00:10:26.200 anywhere but our home. So I brought it here and it's, it's awesome as a conversation starter when

00:10:31.060 people come visit. What I immediately loved about it was when I was 10, so 1983. So just as the,

00:10:40.060 it was kind of the year that the Oilers lost their first Stanley cup to the Islanders who were

00:10:44.660 winning their fourth and I wanted that Jersey for my birthday. And I got that Jersey. It took a lot

00:10:52.240 because the Jersey cost $50. And I just asked my mom if I could ask all the kids at my birthday for

00:11:00.240 money instead of gifts. And I remember everyone showed up with five to $10 and it added up to 50.

00:11:08.320 And we went, I still remember it. Like it's really one of those amazing early memories to go and be able

00:11:12.460 to buy it. And I was really torn between the home and the away Jersey. I went with the home Jersey,

00:11:16.900 which is the one you have. And it was so big. Like I just couldn't, you know, I just wanted to

00:11:23.800 make sure I never grew out of it. And I probably wore it for like every day for three years. The sad

00:11:29.760 thing is I don't have it anymore. I, I, it really bums me out that it somehow got lost because it would

00:11:35.180 be such a cool thing to show my kids and have them roll their eyes at one day. That one's amazing.

00:11:40.160 So wait, you grew up in Toronto and you were an Oilers fan.

00:11:43.920 I was, I mean, the Leafs were not Toronto, such a great hockey town that, you know, going to games

00:11:49.580 was, I mean, just the greatest highlight of my life. And I even got to see a few Oiler games,

00:11:56.160 you know, always in the cheapest seats in the house. Maple Leaf gardens had, you know,

00:11:59.160 sort of this color system. So the gray is up at the very top is where I usually sat.

00:12:03.020 But back then it was so unsophisticated that if the game was a blowout by the third period,

00:12:07.980 you could sneak down into the golds and you could, you know, be like right there. You know,

00:12:12.660 I actually had dinner with Mark Messier a few months ago who I just met and he's just the most

00:12:17.780 amazing guy. And we, I mean, I'm sure he's so tired of doing that, but we just rehashed all of

00:12:23.780 the Oiler glory days. And, you know, it's sort of hard to believe that you could have Gretzky,

00:12:29.420 Messier, Curry, Coffey, Kevin Lowe, Grant Fuhrer all on the ice at one point in time.

00:12:34.800 It was hard to believe before I went to Edmonton, but I took my daughter four years ago now to go

00:12:39.220 to the Women's World Cup. And we went, we decided to go to Edmonton for some reason. I can't remember.

00:12:43.260 It was just like a best chance of seeing the American women's team play. So we spent three

00:12:47.100 days in Edmonton in July and we walked by the old, whatever it was.

00:12:50.660 Northlands Coliseum it was.

00:12:52.020 I think they built a new one. In any case, I thought, wow, it was pretty amazing that these

00:12:56.180 guys would all be here. But then you see Edmonton and you think, oh my God, like I cannot believe

00:13:01.420 they kept them here as long as they did. It's really a remarkably, I don't want to be harsh

00:13:05.800 about Edmonton. I mean, I'm sure there are great things about it, but it's not glamorous. It's

00:13:09.540 certainly not like LA.

00:13:10.220 Yeah. It's not like that was the New York Rangers.

00:13:12.940 Yeah.

00:13:13.400 No. Or even Vancouver.

00:13:14.680 Or Toronto.

00:13:15.060 Or Toronto.

00:13:15.560 Or Montreal.

00:13:16.200 Yeah.

00:13:17.480 Yeah. We share something in common, which I guess is this Hopkins thing. You were a few

00:13:21.020 years ahead of me. So you did med school and residency there, didn't you?

00:13:24.180 I tell people I kind of grew up in the hospital. I wasn't born in Baltimore, but my dad moved.

00:13:29.720 So I was born in Ann Arbor, Michigan, where my dad had done his internship. And like a lot

00:13:34.020 of people at that time and of his age, in a sort of effort to avoid going to Vietnam,

00:13:40.200 he picked up from Ann Arbor and moved to Bethesda, where he worked at the NIH for a few

00:13:44.820 years. It's actually a pretty funny side story, which is I didn't really understand what my dad

00:13:48.440 was doing. I was young. I was only two. And I found a bunch of marijuana cultivation books

00:13:53.340 in his office when I was a kid. And I said, dad, like, is there something I don't know

00:13:58.480 about you? Because you don't strike me as the kind of person who would be. He said, no,

00:14:02.000 this is what I was working on in the lab. We were working on marijuana. He was working

00:14:05.720 in Julie Axelrod's lab. So we ended up living in Bethesda for a couple of years. And eventually

00:14:10.760 he finished and he went back to, he went to Baltimore where he finished his residency

00:14:14.960 and then did his fellowship and joined the faculty at Hopkins in cardiology. And he remains

00:14:20.600 on the faculty there, has been on the faculty for over 40 years. So I grew up going to the

00:14:25.880 hospital as a kid to visit my dad at work. And then my very first summer job was to work

00:14:31.180 in that hospital. So it was sort of a second home for me. And then I went away to college

00:14:34.740 and came back and I did medical school there and stayed around for residency and then moved

00:14:39.000 to San Francisco in 1998, 20, 21 years ago.

00:14:42.060 Did you always want to do medicine?

00:14:43.420 No, never. It was a combination of sort of, I think, not really wanting to do exactly what

00:14:47.300 my dad had done, but also not having the aptitude for it. And I was a pretty bad, I was a very

00:14:51.360 average student in high school, but particularly an average or mediocre science student. And

00:14:56.420 so I went off to college. I went to Vassar, which is a liberal arts school and intended to

00:15:00.300 study music. Actually, I really kind of wasn't going to study music and realized I didn't have

00:15:06.740 the talent to do that either. And just sort of on a whim decided to take a science course.

00:15:12.020 And it was one of these sort of freak accidents that I happened to be at a place where science

00:15:16.220 was friendly and presented in a way that was appetizing to me and non-threatening. And

00:15:20.600 I think if I'd gone to a place like Harvard, there's no way in the world I'd be sitting

00:15:23.860 here right now doing what I'm doing. So I really enjoyed science at that stage in life. And

00:15:29.300 eventually when I figured I wasn't going to be able to do music, I said, you know, maybe

00:15:33.880 I can go off and do medicine, but I'll do something very different. I had no intention of doing

00:15:39.440 science. I had less than the bare minimum science requirements for medical school.

00:15:43.920 So when I got to Hopkins in the fall of 1991, I was probably one of 10 non-hard science majors.

00:15:51.060 And at that time, there were as many biochemistry majors as there were biology majors. It was really

00:15:57.540 unusual to have us humanities people. And we were ranked in the class, you know, one to 120. And

00:16:03.060 we all ended up in the, you know, bottom, we were 110, 111 through 120 and struggled mightily with

00:16:09.720 science, especially with the first year. It was a tough sledding for me.

00:16:13.340 I can't imagine what Hopkins was like then. So I'm obviously a few years behind you. And

00:16:17.840 I got into Hopkins for medical school in 96 or 97. I would have matriculated in 97. I remember

00:16:24.920 even at the time you'd now interviewed at a bunch of schools. So you sort of saw the way it was here

00:16:29.700 versus there. And I was like, wow, this place is going to be rough sledding. I mean, it still had

00:16:35.280 letter grades when most schools had gone to pass fail or high pass pass fail. And you just got the

00:16:41.060 sense like this place was, if it didn't kill you, it would make you stronger, but there was a

00:16:45.220 non-trivial chance it would kill you.

00:16:47.020 It was interesting because of course, you know, from the outside, you have people coming. I was a

00:16:51.340 student member on the admissions committee. And so you got to hear firsthand what people would think.

00:16:55.620 And there was this story that you probably heard, because this was probably around the same time

00:16:59.640 when I was doing it, that Hopkins was this like super intense pressure cooker and everybody was a

00:17:03.760 gunner and you'd never survive. And it was just a miserable place to be a student. It was

00:17:07.840 actually a beautiful place to be a student. And my fellow classmates were really supportive.

00:17:12.840 Everybody was just an incredibly strong achiever on their own. They didn't care how I did. They

00:17:19.040 just wanted to do the best for them. So it was really inspirational. I think after I got over the

00:17:22.620 shock of not being able to keep up, because really the disparity in information was tremendous,

00:17:27.720 right? I'd had one semester of biology and I'm hanging out with guys who, you know, had been

00:17:32.260 working in a lab since they were seven and understood things that like, I still today don't

00:17:37.700 understand. And I mean, it was guys like David Sabatini and Dave Brett and, you know, my high

00:17:44.040 school classmate, Andy Cameron. I mean, a bunch of people who were, you know, just so far ahead of me

00:17:48.500 and where I'd ever be. It was a fun place to be once I got over the idea like, hey, look, I was used

00:17:55.100 to being a pretty good student. And by definition, most of us probably came in here because we were at

00:18:01.180 near the top of our class and it ain't going to happen that way now, right? There are 120 of us

00:18:06.440 and not all 120 of us can get an A. And once I got over that and kind of decided that I was okay

00:18:13.280 with not having all A's unless I wanted to die, it was great.

00:18:17.440 I think I forgot that you were in the same class as David, which of course would put you in the same

00:18:21.520 class as Andy Cameron. And there's an obviously interesting story there is Andy Cameron being the

00:18:25.820 son of John Cameron, who was the chairman of surgery. And one of the main reasons I ultimately did go

00:18:30.580 to Hopkins for my residency, his father, of course, being a luminary in the field. And of course,

00:18:35.860 David Sapatini is no stranger to anybody listening to this. So yeah, it was sort of like, I guess,

00:18:39.440 going to med school at Hopkins in that era was like playing for the Yankees. In the end, I agree 0.98

00:18:43.720 with you. After I interviewed there and I made a point to stay an extra two days, I came away way more

00:18:50.460 bullish on the idea that it could be a great experience. And in the end, it really came down to,

00:18:56.120 I wouldn't say a coin toss. I mean, I think in the end, Stanford won out for a number of reasons.

00:18:59.960 I don't think it was as good a medical school as Hopkins, at least not according to the rankings.

00:19:03.560 Hopkins was the best, but I think just living in California as a kid coming from Toronto seemed

00:19:08.860 like a great opportunity. Well, and I did it the opposite way. I mean, I knew I had to get out

00:19:13.220 after having been there for seven years and having grown up in that environment. And, you know,

00:19:16.720 my dad being there made things complicated. So I knew I had to leave. And it's funny because when I

00:19:21.140 got here, lifestyle is much more central to people here in California. And I didn't appreciate that.

00:19:27.780 And, you know, of course now I do, but I wouldn't trade that experience. It was an amazing experience.

00:19:32.120 And all those people, I mean, I remember operating, I did as a medical student, I did GI gold.

00:19:37.460 You know, that was back when Cameron Yeo and when was Kirk Campbell, the other attending on at that

00:19:42.100 time? No, he, I think he was a chief resident. I'm pretty sure he was. Yeah. So well, I spent a lot

00:19:47.400 of time in the operating room with Dr. Cameron and wow. I mean, what an entertaining human being and,

00:19:52.660 you know, master surgeon, I don't know, were they still doing Sunday school when you were there?

00:19:57.300 Oh yes. In fact, Ted Schaefer and I did an interview a while ago where we talked about

00:20:01.320 our fondness for Sunday school. It was one of the highlights of internship.

00:20:04.220 It was incredible. And as a student, you were terrified because you knew they told you,

00:20:08.900 you know, residents would tell you that you're going to go in there, you're going to present a

00:20:11.940 case. You'll get through seven words, maybe if you're lucky and he's going to cut you off. And then

00:20:16.300 he's going to start asking you questions. So, I mean, this is Sunday morning sitting around this

00:20:20.520 huge boardroom table and you're a little puny medical student in front of, you know, the entire

00:20:25.440 surgical house staff and a lot of faculty. And the most famous surgeon in the United States.

00:20:30.240 Yeah. But of course he had a wonderful way of not humiliating people. I mean, he was rough,

00:20:35.620 but it was an amazing experience. It was really one of my favorite things that I remember about

00:20:40.040 being a medical student back there. I have nothing but fond memories from my time there,

00:20:44.260 despite the challenges. Now, what was a medicine residency like in the nineties? It still must

00:20:48.460 have been pretty freaking rough, especially at Hopkins. So I was an intern in 1996. I think it

00:20:53.140 was the first year they switched from Q3 to Q4 call. And it had been not that long since they

00:20:58.060 switched from Q2 to Q3, but there were still a few rotations where you did every other night call.

00:21:04.100 So as an intern in the medical ICU and the CCU, we did every other night call. And that that's,

00:21:10.580 again, it's one of these things that you look back on, you think, I don't know how I survived it,

00:21:15.100 but in retrospect, it was really fun. But I mean, just imagine what you're doing. You're going,

00:21:18.440 you start rounds at whatever, eight o'clock in the morning, you go all day. At that time,

00:21:23.260 you never left the unit. So people would bring you food. There was a bathroom and a call room

00:21:28.400 adjacent to the unit. You never left the unit. And then you'd stay up all night taking care of

00:21:33.340 patients. You'd get to rounds the next day, you'd finish by 11 o'clock. And we had this tradition

00:21:36.920 of going out to have breakfast and beer with the ICU nurses. And then you'd go home, you'd take a little

00:21:42.600 nap, you wake up, you'd have some dinner, you'd go back to sleep, you'd do the whole thing all over

00:21:46.280 again. So for 30 straight days, you basically did nothing but sleep or be in the hospital.

00:21:50.280 But it was fun. It was an incredible experience. I mean, it was still, there were still a lot of

00:21:54.560 active HIV disease. And one of my favorite inpatient rotations was doing what they used

00:21:59.760 to call the OCR8 rotation, which is the inpatient HIV ward. And I mean, you did that as a junior

00:22:05.140 resident with two other junior residents and there was no one else around. And it's basically ICU level

00:22:10.360 care. I mean, these were some of the sickest people in the world. This is, you know,

00:22:13.120 pre-protease inhibitors. So sick, sick, sick people, but amazing. I mean, I remember doing

00:22:18.060 procedures on people who are still my dear friends, fellow residents, really an amazing

00:22:22.180 experience. So how did you pick cardiology? Well, again, so the background is I'm a second

00:22:26.420 generation cardiologist. So my intention was to avoid it for a lot of reasons, but I also just wasn't

00:22:32.500 really drawn to it. But actually I did a CCU rotation as a, as a medical student. And I'll never

00:22:37.540 forget, we had a series of people, patients come in who were kind of in their late thirties,

00:22:42.580 early forties, who had myocardial infarction, had a heart attack and had relatively absent any

00:22:48.660 obvious risk factors. So our assessment of risk factors for coronary disease hasn't changed a lot

00:22:54.540 in the past 20, 25 years. But what I'm saying is that they had normal lipids, normal blood pressure,

00:22:59.660 nothing identifiable. And I thought, wow, this is really kind of fascinating. So I worked the summer

00:23:04.940 between first and second year. I worked in a lab. I worked in a neurosciences lab. I was working

00:23:09.060 on the genetics of trinucleotide repeat diseases. It hadn't been proven yet, but there was a strong

00:23:13.860 suspicion that some of these psychiatric diseases like schizophrenia and bipolar depression were

00:23:19.880 mediated by trinucleotide repeats. And it was my first time in a lab and I loved it. So I knew I

00:23:25.220 wanted to go back and do that. I ended up basically asking a bunch of people, I said, you know, is anybody

00:23:29.780 here doing work on the pathogenesis of, it wasn't called ACS or acute coronary syndrome back then,

00:23:35.740 but the pathogenesis of unstable angina or MI in young people without risk factors.

00:23:40.680 So I got pointed to this guy who was a young, at that time, assistant professor, Belgian guy,

00:23:45.400 who went on later to be the dean of the medical school at the University of Miami. He had a

00:23:49.840 blustrious academic career. His name is Pascal Goldschmidt. And so I met with Pascal and I said,

00:23:54.380 Pascal, I'd love to come spend a year working in your lab. And he said, great. So that sort of was

00:24:00.340 my next introduction to science. And it was really that experience, which was phenomenal and career

00:24:05.860 defining for me that drove my clinical interest more than the other way around. So I kind of at

00:24:10.000 that point had thought, well, this is kind of fun, but maybe I'll do adult cardiology or maybe I'll do

00:24:14.500 pediatric cardiology. And I'm not entirely convinced I know the answer yet, but I ended up eventually

00:24:20.120 deciding that I didn't like parents. And so I went down the path of doing adult cardiology.

00:24:25.200 I want to touch on all the stuff you just brought up because you and I have had so many discussions

00:24:30.160 about this over the past couple of years. And I've always found you to be one of the most lucid

00:24:35.760 people at describing these distinctions and sort of synthesizing this. And for someone like me,

00:24:41.080 who is not really on the front lines of acute cardiology problems, it can be an overwhelming

00:24:47.580 body of literature. So let's just start with some of the semantics, acute coronary syndrome,

00:24:54.380 unstable angina, stable angina. How would you define all these terms for folks? And what's

00:24:58.600 the framework in which you anchor these things? This is what's amazing is if you go back in time,

00:25:03.280 it was relatively recent in history that there was a debate about the cause of most myocardial

00:25:09.800 infarctions. 99% of heart attacks in this country happened because of a ruptured atherosclerotic plaque.

00:25:16.620 And that's worth even pausing on for a moment because I don't even think that most people realize

00:25:20.220 that. I think most people think that a heart attack occurs when the pipe narrows, narrows,

00:25:27.160 narrows, narrows, and finally closes. So that when you hear about so-and-so who has a 95% occlusion in

00:25:34.400 the artery on the angiogram, my God, that's just 5% away from killing them. And they sort of don't

00:25:39.860 understand what I think you just said. So can you double click on that point?

00:25:43.400 Yeah. Well, this is what's so amazing is that if you go back, not that far in the literature,

00:25:47.000 you'll find articles where people are having a debate about whether the, so they, they did autopsies

00:25:52.700 on people who presented with sudden cardiac death. So somebody would show up dead from a heart attack.

00:25:57.080 They do an autopsy, they'd open up the heart, open up the artery, and they'd find a clot there. And

00:26:01.520 there was a debate about whether that clot was primary or secondary. Was it there before or after

00:26:05.060 the person died? Was it the basis? And it really was the work of Michael Davies in the late 1970s and

00:26:10.200 early 1980s that kind of demonstrated, proved the idea that this was actually causal. And then

00:26:15.800 early 1980s, I think it was, I can't remember the exact trial. They went in and actually aspirated out

00:26:20.660 clots and people who showed up in the emergency department with, with a heart attack and those

00:26:25.240 people got better. So this was pre-angioplasty, but they were able to aspirate out the clot using a

00:26:30.660 suction catheter and people got better. And so that was the first demonstration that this clot event

00:26:36.280 was the primary cause of most heart attacks. So again, the pathologist leading the discussion

00:26:42.100 on how that happened and what you could see in anybody was you'd see a lot of these plaques

00:26:47.720 and Davies described, I think, I can't remember which paper it was. I'll have to go back and find

00:26:52.060 it for you. But he described something that stuck with me, which was that if you look at an autopsy

00:26:56.640 of somebody who presents with a, you know, with a heart attack and dies, you'll find, you can tell by

00:27:02.040 looking at the histology, how many times the plaque is actually ruptured, much like you'd look at the

00:27:06.660 age of a tree. So you can see that there's plaque rupture and then there's healing. And actually now,

00:27:11.380 you know, other pathologists like Reno Varami and others have said that a lot of the progression of

00:27:14.940 coronary disease. So getting narrower and narrower is probably that event happening sequentially. So

00:27:19.420 plaque rupture, healing. And then as you do that, you're laying down more fibrous tissue. So he could

00:27:24.100 see how many times, and on average, these people in that first series had had seven plaque ruptures in

00:27:30.220 their culprit artery before they died. At the same point in the artery or along different points?

00:27:35.900 In the same plaque. So in the same plaque. And so that told-

00:27:38.480 So basically a seven ringed tree. On the cover of Starry's pathology book, which is,

00:27:43.840 I've talked about this before, it's sort of one of the most important books I have to explain

00:27:47.480 atherosclerosis to patients. There's actually a picture on the cover that shows a very comparable

00:27:51.940 lesion to that, which is just these concentric rings where the final one, you have this big,

00:27:57.400 dark, bloody clot in there, which was obviously the fatal one.

00:28:00.340 That was what lit the fire for me scientifically back when I went to Pascal. I was struck with this

00:28:06.380 idea, well, gosh, if it, like, why didn't somebody die at the first? Why don't they die after the

00:28:10.740 first one? Like if the reaction is, so just for listeners, the contents of that plaque, what we're 0.95

00:28:16.040 calling this atherosclerotic plaque, it's rich with lipids and inflammatory cells. But the thing that

00:28:22.180 causes the heart attack is none of that stuff. It's that the macrophages, which are these white blood cells,

00:28:27.000 carry around this protein called tissue factor, which is then basically the instigator or the

00:28:33.120 trigger for the clotting cascade. What you're doing is exposing blood, flowing blood to tissue

00:28:38.300 factor, which is basically a wound healing signal. So if you cut yourself, we have tissue factor

00:28:44.200 underneath the endothelium, which is that inner lying of our blood vessel. And that's there

00:28:47.680 to sense injury. And when the blood sees tissue factor, the blood, there's a biochemical reaction

00:28:53.440 that we've been described for decades that ends up resulting in a blood clot.

00:28:57.700 It's sort of an epoxy, you know, those epoxies that come in two separate tubes taken alone.

00:29:02.120 They're not sticky, but when the tube combines them, it becomes like rock solid glue. So the

00:29:07.140 macrophage exposes the tissue factor. Now the blood passing through becomes the other tube of the epoxy

00:29:12.040 and boom, you get this acute brick. I've been describing it as a substrate and trigger, but I think

00:29:16.900 that's a much better, I like your description. So, and that was again, a debate, not that long ago

00:29:22.600 about what the cause of these things was. Eventually within a 10 year period, we learned so

00:29:27.800 much because it was, I think 1988 that ISIS-2 was published and ISIS-2 was the, this is off the top

00:29:32.960 of my head, but I think it was a comparison of aspirin and streptokinase versus aspirin alone in

00:29:38.800 patients who presented with heart attack.

00:29:40.400 Tell folks what streptokinase does.

00:29:42.420 Streptokinase is a, what's called a thrombolytic. So streptokinase is a natural substance that we make,

00:29:48.740 or we make something similar that dissolves clots. So clotting is sort of this balance,

00:29:53.320 this, this battle between pro forces and anti forces. Because if you think about it, if you're

00:29:58.700 a mammal and you have a circulatory system, you have to be able to sense injuries so you can keep

00:30:03.240 your blood from flowing out of you. So the system has to be incredibly tuned and you as an engineer

00:30:08.400 will appreciate how you must be able to respond incredibly quickly, but it has to be specific.

00:30:12.520 So you can't get clots happening. We all know what happens.

00:30:14.980 That's such a great point to make, right? So let's go back and think about this through the lens of

00:30:17.700 evolution. You get a scratch across you as bad as that scratch or cut is, it still represents a

00:30:23.820 fraction of the surface area of your vascular system. So to your point, you have to have a

00:30:29.760 system in place that can identify the area that needs to be fixed. Hence the two check system where

00:30:36.000 I have to have the tissue factor, which allows for the local aggregation. And then very quickly you

00:30:41.480 have to respond. And yeah, if it was just a diffuse response, well, you'd kill the organism by

00:30:46.340 clotting everything within the circulatory system.

00:30:48.720 Well, that's right. And we really appreciated that through human genetics and then later mouse

00:30:52.600 genetics. So if you knock out any of the pro coagulant coagulation factors, say, you know,

00:30:59.440 any of the one, you know, factor two prothrombin, you'll get a sick mouse, but the mouse will survive

00:31:04.440 through development. So they will, they will survive. And then what kills them is the process of

00:31:08.640 being born. So they'll die right after birth. If you knock out one of the anticoagulants and knock

00:31:13.100 out, say protein C, protein S, any of the anticoagulant, the mouse won't make it through

00:31:18.620 half of the embryonic, basically day seven or day eight, they die. And then there's the whole

00:31:23.920 other interesting layer, which we won't have time to get into today, which is the role of the placenta

00:31:27.880 and how placentas don't clot off. And there's just unbelievable, fascinating evolutionary biology that

00:31:33.400 unless someone has help syndrome, that becomes this awful exception to that rule where you get

00:31:39.060 these infarcted, clotted placentas, and then the fetus is usually underdeveloped. So.

00:31:44.840 Well, I'm not an expert, but I think that one of the most common causes of fetal loss of

00:31:48.700 idiopathic fetal loss. So women who end up losing multiple pregnancies again and again and again

00:31:53.300 is an undiagnosed hypercoagulable state, I think, because basically the placenta ends up clotting off.

00:31:59.340 Yeah. You're saying in the case where you don't have an obvious aneuploidy that,

00:32:02.780 or a chromosomal abnormality. Yeah. So is it easier to talk about where we were 10 years ago,

00:32:08.300 where we are today, or do you want to just define today? So you've talked about what a

00:32:11.500 myocardial infarction is.

00:32:12.820 Yeah. We got sidetracked.

00:32:14.220 So that's what I do here. I sidetracked.

00:32:15.820 That's all right. That's fun. So myocardial infarction was defined clinically as most things

00:32:20.280 were and not pathophysiologically until recently. It was defined by the constellation of things that

00:32:25.640 still, I think, make up most of the definition, which is chest pain, changes in the electrocardiogram

00:32:29.900 and presence of biomarkers. Let's explain what those three things are. So everybody

00:32:33.660 understands what chest pain is. You have these classic symptoms that people understand,

00:32:38.160 elephant sitting on chest, radiating into left arm, nausea, that sort of stuff up into the neck.

00:32:43.240 You mentioned the changes in the electrocardiogram. We'll obviously link to photos of this sort of

00:32:50.280 stuff, but it's hard to do this because we're going to try to do it for the listener. But every time

00:32:55.220 the heart beats, there's a signature that can be measured electronically when you put leads on the

00:33:00.080 chest and you have a little wave that symbolizes when the atrium is doing its contraction. That's

00:33:05.120 called the P wave. Then you have the big signal, which is this QRS spike, and that's this ventricular

00:33:11.580 contraction. And then you have another little wave called a T wave. And I can't even remember.

00:33:16.420 Is that the repolarization?

00:33:17.760 Yeah.

00:33:18.160 Ventricular repolarization.

00:33:18.880 Ventricular repolarization.

00:33:19.640 And then it goes again. So what are the signs on that, that you are looking at when you're in

00:33:25.220 the ER and somebody comes in to help you decide if this person is having a heart attack or maybe

00:33:31.000 even has had one in the past?

00:33:32.600 Yeah. It's important that we cover this now because actually after it was agreed upon that

00:33:36.300 a ruptured plaque led to a clot, the clot led to basically the absence of any blood flow distal to

00:33:41.700 that clot. So beyond the clot. So then any tissue beyond that blockage would then be starved of oxygen.

00:33:47.620 And after some period of time, a few minutes, five minutes would die. And wasn't that long ago,

00:33:52.860 the treatment of MI of heart attack was to put somebody in bed. And then there was the advent of

00:33:58.120 beta blockers, a couple other blood pressure medicines, which were basically forced, you know,

00:34:03.080 pharmacological bed rest and then aspirin and then streptokinase, which we talked about,

00:34:07.080 which is basically a clot dissolving medicine. That trial demonstrated that if you dissolve the clot

00:34:12.120 pharmacologically, you could rescue people. And then we got into actually going and doing these things

00:34:17.540 with balloons and later with stents. But the nomenclature at the time that I was a medical

00:34:22.580 student was you defined myocardial infarction. So MI by the presence or absence of the Q wave.

00:34:27.880 So you mentioned on the EKG, there was this big deflection that happens early on and that represents

00:34:33.820 depolarization of the ventricle. So the biggest, largest mass in the heart is your left ventricle.

00:34:40.160 So that's the pumping chamber that pumps blood out to the body. And that has a very strong electrical

00:34:45.240 signal. And if in certain areas, regions on the EKG, so we do 12 leads and it creates a vector and

00:34:53.060 we can basically recreate the heart in three dimensions. And so you're doing that by putting

00:34:57.360 these leads, these electrodes across the chest and on the arms and the legs. And if after somebody had

00:35:02.880 a heart attack, they developed a Q wave in certain leads that signaled to us clinicians that they had

00:35:10.300 what we at that time called a Q wave myocardial infarction. And what that meant was that there

00:35:14.740 was transmural or all the way through cell death. So the entire wall, there had been an actual

00:35:20.820 necrotic, basically we had become necrotic and died. Whereas there was another version of heart

00:35:26.520 attack at that time that we called non Q wave heart attack. And those people presented with the same

00:35:32.580 clinical picture. So they had chest pain. They also had positive biomarkers. At that time,

00:35:39.440 we were measuring CK or creatine kinase. And that's an enzyme that's leaked out of the

00:35:44.300 cytosol of a muscle cell when there's damage to it. And so people have heart attacks. There

00:35:48.520 are a bunch of these enzymes, myoglobin and others that go up. And that's one of the ways that we

00:35:52.880 define heart attacks. So people would come in, they'd have a positive biomarker, they'd have chest

00:35:56.000 pain, but they didn't have this Q wave. And so they were called non Q wave myocardial infarction.

00:36:00.820 So basically the diagnosis was based on the chest pain and the leaking of the enzymes. So I'll say a

00:36:06.600 little bit more about that because we're going to come back to it. But basically when a muscle dies,

00:36:12.000 you can pick up signs of muscle death in the blood and CK being one of them. In fact, if someone goes

00:36:19.440 out and runs a marathon, it's not uncommon that the next day, if you draw their CK level, it's very

00:36:25.220 elevated. They've broken down muscle. My recollection, because it's been so long since I've looked at

00:36:30.100 CK is they used to even look at a fraction called the MB fraction, which would try to be more specific

00:36:36.480 for cardiac muscle than say the muscles in your quads or something like that. So if you had those

00:36:42.560 two things, the pain and the enzyme findings, you were having an MI, the EKG was simply used to

00:36:48.720 determine if it was Q wave completely through or non Q wave. Is that right? Mostly. I mean,

00:36:54.660 the EKG is fundamental. I mean, it's really the first thing that gets done when somebody shows up,

00:36:59.000 they're actually doing them in the field now. And it does today define how we treat people. It

00:37:03.800 defined it as well back then, but things have changed significantly in the past 20 years. And

00:37:07.300 we can touch on that a little bit, but the EKG is, is a fundamental tool. We can't get away without

00:37:13.040 it. So that was the way it was described. And then there was this other thing called unstable angina

00:37:16.840 and unstable angina was thought to be the same pathophysiology. So ruptured plaque,

00:37:21.280 but it was thought to be incomplete cessation of blood. So maybe the clot didn't completely block the

00:37:26.400 artery. So there were still a little bit of a blood flow getting through. So in that case,

00:37:31.300 back then, and again, it's changed now, but back then, if you had chest pain and the EKG changes,

00:37:37.620 but you didn't have a prize in the biomarker, you were said to have unstable angina.

00:37:43.200 And that was differentiated from stable angina based on what?

00:37:46.800 Yes. So stable angina has been described forever, which is you walk and you get these symptoms that

00:37:52.900 people describe and we can come back and talk about what people, we call it chest pain. That's

00:37:58.160 a garbage basket term that comprises a lot of different things ranging from exercise intolerance

00:38:03.800 to actually that feeling of, of, you know, Fred Sanford talking about, I'm having the big one or

00:38:07.800 an elephant sitting on your chest. That is a clinical term of art, basically the way I think about it.

00:38:13.000 So chest pain presence or absence of the biomarker and the EKG. And that would define the,

00:38:18.880 the symptoms. Stable angina was defined as you're exercising, you're increasing demand. So again,

00:38:26.000 what we're talking about is an imbalance between myocardial oxygen supply and demand. So what used

00:38:31.860 to be called stable angina and still mostly called stable angina was thought to be a demand issue.

00:38:36.720 So there would be a lesion that would narrow the artery and you'd only get chest pain when you

00:38:41.180 increase demand. Sometimes people call it demand ischemia. And so the idea was that if you have this

00:38:45.720 blockage that's 75, 80% of the diameter of your artery and you go out and you do something that

00:38:51.460 causes you to increase your demand. So you go walk up a flight of stairs or a hill here in San

00:38:55.760 Francisco, you need to be able to deliver more oxygen to the tissue and you can't do that. So the

00:39:00.000 way the coronaries work physiologically is that they augment blood flow by dilating. So it's a

00:39:04.800 different physiology than most other vascular beds. So it's part of the reason these drugs like

00:39:10.080 adenosine work is that they dilate the coronary.

00:39:12.720 That's why people are told to take nitroglycerin if they're having chest pain.

00:39:15.940 That's right. But the problem is if you've got a fixed lesion there, that's full of a calcified

00:39:19.960 plaque with a bunch of different lipids and stuff in there, you can't really dilate. You can dilate

00:39:23.940 the proximal segment before the blockage, or you can dilate the distal segment after, but you can't

00:39:28.040 really deliver more. And so you end up getting an increase in blood flow to all the other segments.

00:39:32.920 And then you get a little bit more of what probably a little bit of coronary steel. So in other words,

00:39:36.760 you're getting more blood is even being taken away. And so that syndrome is considered to be

00:39:42.500 stable angina. So you're walking up a hill, you get chest pain, shortness of breath, whatever

00:39:47.000 the symptom is, you sit down and you rest and it goes away. Or like you mentioned,

00:39:51.800 you take a nitroglycerin and it goes away and that's stable.

00:39:54.960 That's actually the best explanation I've ever heard for it. Understanding the fact that I'm not

00:39:58.360 a cardiologist. And even when I was learning this stuff in medical school, maybe I just wasn't paying

00:40:02.760 close enough attention, but I think the way you just explained it is great. It's not just that in the

00:40:07.580 unstable case, you may have some actual plaque rupture. It's this idea that in the stable case,

00:40:13.160 it's the inability to dilate the target lesion and the plausible steel of blood. So morphologically,

00:40:21.460 it's a totally reversible process. Nothing has changed. There hasn't been a new injury,

00:40:26.840 but under a certain period of stress, you're able to basically see this area that otherwise can't

00:40:34.000 function fully. That ties in together with sort of the way symptoms develop, the chronicity of the

00:40:39.000 problem, the fact that people talk about how stable engine doesn't typically kill people. It develops

00:40:43.820 slowly over time, months or years. There can be an acute process that overlays that, but for the most

00:40:50.040 part, it's slow. And that is the thing that we spend the lion's share of our time thinking about and

00:40:56.320 talking about and treating, but really it doesn't kill you. What kills you is the acute thing. So just to

00:41:02.280 kind of put this in a little bit of reference for you. So having coronary disease, which most

00:41:08.080 Western adults will have by the time they get to be 50, meaning if you looked by CT at me or you,

00:41:14.380 maybe not you, but I'll probably have a little bit of plaque, but it's very unlikely that that plaque,

00:41:19.280 the amount of plaque is obstructing blood flow, either significantly at rest or with exertion.

00:41:25.060 Over time, if that plaque gets bigger at some point, it'll start to become apparent to me

00:41:31.140 that when I'm walking, increasing my demand that I'm not able to deliver oxygen. So I'll become

00:41:36.260 symptomatic. And that the point at which we say most people become symptomatic is when the

00:41:41.520 percent diameter stenosis or blockage is about 70%. Conveniently, it's also the point when stress

00:41:47.860 tests, whatever flavor you like, also begin to be able to detect coronary disease. So we have focused,

00:41:55.740 I think a lot of energy on this 70% number because it's what we can measure clinically,

00:42:00.860 either through symptoms or through our testing, but there's nothing to suggest that there's a magic

00:42:06.340 in that number that would put somebody at a greater risk as opposed to somebody who had a 50%. And in

00:42:13.000 fact, if you go back over time and again, to the work of Michael Davies, I think, and I don't want to

00:42:18.020 misspeak here, but I think a lot of the fatal plaque rupture events that happened in this series of people

00:42:25.740 in arteries that had only 30% or 40%. I think that's actually correct. And I mean, that's sort

00:42:30.600 of one of the things I remember from that literature that was very surprising, which was, it was very

00:42:37.020 difficult, a priority to predict which one of these things was going to be the fatal one. You could have

00:42:41.820 these 90% occlusions that were highly symptomatic, but just incredibly stable. And they were never going

00:42:48.000 to rupture and they were never going to be the things that killed you. And in fact, there's almost a

00:42:51.120 paradox, right? Which is the more the stenosis increases, the more likely it tells you that

00:42:57.280 the downstream musculature has either found some other way to acquire it. So, so, you know, someone

00:43:02.680 who's walking around with a 95% stenosis is very likely to suffer death in the case of that plaque

00:43:10.400 rupturing. Of course, that's still a harbinger for what's going on elsewhere in the heart. So there

00:43:14.740 obviously, you know, very few people are walking around with a single point of this injury. It's telling

00:43:19.800 you that there's probably injury throughout the heart and that they could die from another one. But

00:43:23.380 that to me, which I want to come back to, right? Because I want to talk in depth about some of the

00:43:28.760 diagnostic tools that we have, especially to predict who's at risk, who's not at risk. And this

00:43:33.960 is becoming increasingly more important as we have to decide both on the risks and costs of treatment

00:43:42.280 as the treatments get more and more elaborate. And this ability to predict who's going to have a

00:43:48.740 fatal MI versus who's not. I, I'm not convinced we're that much better at it today than 20 years

00:43:53.640 ago. Are we? I don't think so. I mean, I think we should definitely come back to that. That's a

00:43:56.920 fascinating discussion. We could spend eight hours talking about that. Yeah. Let's go to where we

00:44:01.060 are today. So we've talked about Q wave, non-Q wave. Talk to me about what is the nomenclature

00:44:04.180 today? This nomenclature evolved over the past 15 or 20 years. And the reason it did was that it was

00:44:09.720 sort of the advent of interventional cardiology. And what people realized was in the acute setting of a heart

00:44:15.420 attack. So I was talking about that Q wave before that appears days after a heart attack,

00:44:19.960 but in the acute setting of a heart attack, there was a defining feature that seemed to predict whether

00:44:25.180 there was a complete loss of blood flow. And that was something that we describe as called ST elevation.

00:44:30.120 So you described beautifully the different waves. And so it's the QRS complex. So the end of that big

00:44:35.240 squiggly complex, the segment, there's then a line, a direct line, a flat line that connects the end of

00:44:41.680 the S wave to the beginning of the T wave. And that we call the ST segment. And that is usually

00:44:46.600 isoelectric with the rest of the EKG. In other words, it's at the same level as the PR segment

00:44:54.040 and everything else. What people realized was that if that segment gets increased above a couple of

00:45:01.540 millimeters, and it does so in more than one lead, that that signaled that there was a block,

00:45:05.940 a complete blockage of an artery. And that then evolved into this term that we use today still,

00:45:11.420 which is called ST elevation MI or STEMI. And that is a medical emergency immediately signal,

00:45:18.880 we need to open this artery. And that's work that was done by Gene Roundwald and other people in the

00:45:23.560 TIMI group. And it was done with a combination of pharmacology. So streptokinase, but then also

00:45:28.900 TPA and other thrombolytic agents. And then people realized you could actually go in there and open the

00:45:33.880 artery manually with a balloon catheter. So the standard of care in this country today is that if you show up

00:45:39.680 with ST elevation and a clinical picture that looks like MI, you end up going to a cath lab within

00:45:45.960 an hour and you get that artery opened up. What is the overlap between ST elevation? And if you go

00:45:53.320 back and look at all the EKGs that define Q wave and non-Q wave, I'm guessing the non-Q waves rarely

00:45:59.020 had ST elevation and some of the Q waves, but not all of them did.

00:46:02.680 Yeah, I think that's the right thing. So the way to think about it is the Q wave and 0.90

00:46:06.260 the ST elevation MI are probably the same. It's just, you're seeing a later manifestation of the

00:46:10.040 Q wave. If somebody had showed up and if at that time they did EKGs on everybody who walked into the

00:46:15.180 emergency room with chest pain, like they do now, you probably would have seen the ST elevation.

00:46:19.100 They probably just missed it. So they saw the Q wave because that was what came later.

00:46:23.280 The non-Q wave MI is probably now what we call non-ST elevation MI. And it's actually evolved to a

00:46:29.660 different term, which is, I don't love, which is called non-ST elevation acute coronary syndrome.

00:46:34.080 Which is just not as easy to say as non-ST.

00:46:36.260 No, no. So acute coronary syndrome, I think the way I tell the, teach the students is acute

00:46:41.080 coronary syndrome is the entire collection. It's everything. And it includes STEMIs, non-STEMIs,

00:46:46.860 everything. And it includes what we used to call unstable angina. Unstable angina doesn't really

00:46:51.260 exist anymore. So we now have ST elevation ACS or ST elevation MI, and we have non-ST elevation ACS.

00:46:59.180 And that's very simply defined by the presence or absence of ST elevation.

00:47:02.900 Turns out that people who don't have ST elevation often have other ST segment abnormalities.

00:47:08.260 Most commonly their ST segments are actually depressed. And that has to do with changes

00:47:12.560 in the repolarization. I'm not an electrophysiologist, but changes in the

00:47:15.760 repolarization that happen when the muscle is completely starved of oxygen versus only partially

00:47:20.600 starved. It's a super interesting thing that I've never understood. I probably should go

00:47:24.240 learn why that is. So clinically, if somebody shows up in the emergency room with ST elevation,

00:47:29.220 it triggers 911, activate the cath lab. This patient goes straight in, they get heparin,

00:47:34.700 they get aspirin, they get all the other things, clopidogrel.

00:47:37.340 And just to be clear, do we even check enzymes in those patients or we don't care?

00:47:41.080 No. So it's an immediate thing. And in fact, when I, and this change happened during the time I was a

00:47:45.520 cardiology fellow. So when I was a cardiology fellow, somebody showed up in the emergency

00:47:48.460 department with ST elevation, they'd call me and I'd come down there and I'd look at the EKG and then

00:47:53.160 I would make the decision to activate or not activate the cath lab. But since Gene Brownwell

00:47:58.060 coined this term, time is myocardium, since we understand that the longer that the heart is

00:48:03.580 deprived of oxygen, the less chance that it's going to actually recover, there's been a push to speed

00:48:08.640 things up. And so now there's...

00:48:09.840 So the ER can now make that call.

00:48:11.900 Yeah. It's actually often made in the field. So the ER makes the call and the page goes directly to

00:48:17.240 the interventional cardiology team and the cardiology fellow and the attending. So when I'm

00:48:21.200 on service, I often learn about, I'm the last one to hear about a patient of ours who's come in

00:48:24.920 with a STEMI.

00:48:26.320 You hear about it when they're in the cath lab or maybe on the way out.

00:48:29.480 Yeah. I hear about it when the student or the intern calls and says, what do you want this dose

00:48:33.340 of this to be? In the old days, wasn't that long ago, 15 years ago, this cardiology fellow had to

00:48:38.800 make that call.

00:48:39.400 How much time has that saved? If you think about where we are today versus 10 years ago,

00:48:42.980 that's a half an hour savings potentially.

00:48:44.940 Maybe. I mean, every hospital now in this country is measuring their what's called door

00:48:48.660 to balloon time or door to open artery time. It's a quality metric. And like every quality

00:48:53.780 metric, there are trade-offs, right? So there are more people taken to the cath lab who probably

00:48:58.860 don't belong there. There are some people taken there who are maybe DNR, DNI, or have metastatic

00:49:03.840 cancer, or there are other things that you don't have the time to have a nuanced conversation

00:49:07.900 and you're not having a conversation with an expert. So there is a trade-off, but the overall

00:49:13.320 net-net, I think there's our society of people, citizens and cardiologists and other people

00:49:18.480 and emergency room doctors, I think all agree that this is better than it used to be.

00:49:22.320 The dilly-dallying that would happen and the delays are gone. So the most important thing

00:49:26.840 is get somebody up there. If it's a mistake, it's a mistake. It's not the end of the world.

00:49:30.240 It's a little bit of radiation.

00:49:31.980 So which trial, was it courage that made this case? Which is the trial that told us

00:49:36.440 that what you just said was really the right way to do things, that opening the artery in

00:49:40.940 the STEMI patients is going to save lives?

00:49:42.660 I'd have to go back. It was probably one of the early TIMI trials. I don't remember which one. I'd

00:49:46.680 have to go back and look about which one. There were a series of trials done in the 1990s comparing

00:49:51.220 what used to be called primary angioplasty. So angioplasty is the Greek way of saying opening an

00:49:57.720 artery with a balloon.

00:49:58.420 Can you explain just, I mean, there's going to be a lot of people listening to this who

00:50:01.040 won't necessarily know exactly what we mean. So when a patient goes into a cath lab,

00:50:05.940 what actually happens?

00:50:07.980 So again, a little bit different today than it was in the old days, because in the old days,

00:50:11.820 almost everybody had basically a needle put into the femoral artery. So the femoral artery is a

00:50:16.180 major offshoot of the aorta, feeds your leg oxygen. 0.81

00:50:19.920 You go into the crease in the groin and you can, anybody can feel their pulse there. So if you feel

00:50:25.400 that, that's your femoral artery. That's about the size of your, you know, somewhere between your

00:50:30.060 pinky and your middle finger. I mean, that's a monster artery and you're putting a catheter in

00:50:35.180 there.

00:50:35.800 There's a technique called the modified cell ginger technique, which allows you to put a catheter

00:50:39.200 into that artery. So that catheter then allows you to have access to the artery. At the tip of

00:50:44.200 that catheter is a one-way valve. You put things in, but blood can't come back out. So through the

00:50:48.660 tip of that catheter, there's a little plastic one-way valve and you can thread things in. You can

00:50:53.740 thread in other plastic catheters. So in this case, you thread in a regular coronary catheter,

00:50:59.060 which I think is like two millimeters. You thread that in. You can then under x-ray guidance

00:51:03.660 with a wire to make it a little stiffer, guide it all the way up around the arch of the aorta and

00:51:08.820 back down. We'll have a figure of this for sure. So people can see what that does back down to the

00:51:13.680 base of the heart where the coronary osteo are. So they're normal human beings. There's a left

00:51:18.560 coronary osteo and a right coronary osteo, and you can put that catheter and engage it. And then

00:51:23.860 that allows you to inject contrast dye, which under x-ray will allow you to actually see the artery or

00:51:30.440 see at least the inverse of what the artery looks like. And then what people realized was you can

00:51:35.380 deliver other things through these catheters too. You can deliver another smaller catheter. And on the

00:51:39.980 tip of that catheter, you could wrap a balloon and then basically you could use a syringe to blow up that

00:51:45.660 balloon. And that would basically crush open the blockage or clot or combination of whatever else it is

00:51:52.940 that's there and reestablish blood flow.

00:51:55.620 So it's just, it's a very mechanical process. I mean, when people, I don't think necessarily people

00:52:00.180 appreciate just how mechanical and in some ways crude this is at the, you know, you described an

00:52:06.600 incredibly complex chemical reaction that is taking place. And one of our solutions is chemical, right?

00:52:13.220 We can inject something that can break apart the epoxy, but also we can put a balloon in and use that

00:52:19.360 balloon to inflate open and try to crack it open. And as you'll mention in a moment, I'm sure you can put

00:52:24.840 in these metallic things that spring open called stents.

00:52:28.600 It's such a good point, Peter. I mean, it's really, it's as if you went to a stream and there were a

00:52:33.720 beaver dam there and you just take a bulldozer, you take a shovel and just shove all of it downstream.

00:52:39.320 And so, you know, one of the questions I've always asked is, well, why is this therapy not more effective?

00:52:45.140 And we'll get into where it's effective and not effective, but is there potentially a downside to 0.62

00:52:49.560 elaborating all of that crap downstream? And does that have an effect on smaller blood vessels or

00:52:54.260 smaller areas of muscle that end up getting occluded by all this stuff that you elaborate downstream when

00:52:59.220 you're blowing open this artery? And actually goes back to the question of sort of when people were

00:53:04.160 doing these trials to compare the efficacy of what's called primary angioplasty versus

00:53:09.360 thrombolytics, TPA or streptokinase, these chemical drugs.

00:53:13.220 So the mechanical versus the chemical.

00:53:15.260 That's right. The reason that those drugs are not more effective is not that they're not

00:53:20.260 great at dissolving blood clots, although they're not perfect at it, but that by doing so,

00:53:25.740 you're increasing the risk of dislodging a blood clot somewhere else that might be really important.

00:53:29.740 So let's just say you'd had an injury in your head and there was a blood clot somewhere in your

00:53:34.400 subdural area. And so if you go in and you blow in this TPA, you're basically dissolving all blood

00:53:39.380 clots and it's, it's not precise. It's sort of a imprecise, just every blood clot. So if you have

00:53:44.660 one somewhere in your GI tract, you'll start bleeding out there. And so the consequences of

00:53:49.540 doing that, I think we're great. And that's part of the reason why primary angioplasty again became

00:53:53.160 standard of care in this country.

00:53:54.700 Because while the epoxy process requires two things and therefore is highly specific and be

00:54:00.640 localized, the clot busting is not. You have to put that solution into the circulation. And even

00:54:07.200 though you apply it locally, every time that heart beats, it gets dissipated. So every part of the body

00:54:12.820 is going to see it.

00:54:13.660 So there are certain circumstances where people try to deliver TPA or cousins of TPA locally. Like I think

00:54:19.500 even in stroke, they're still injecting TPA and trying to keep it relatively local, but it's a

00:54:24.760 very difficult process. Cause as you say, circulation is circulation, right? I mean, it doesn't take that

00:54:28.720 long for something that's part of this part of your circulation to get around your whole body and come

00:54:34.480 back again. So that's why I think, well, I should say when I was a cardiology fellow in 2001, 2002,

00:54:40.400 we were still doing using TPA as a way of treating people who showed up with a heart attack at San

00:54:45.480 Francisco General Hospital. That's not the case anymore. I think most, almost all U S hospitals

00:54:50.960 now it's primary angioplasty. And so after you mentioned these stents, these little metal scaffolds,

00:54:57.400 they developed back in the 1990s as a means of treating stable angina. So if somebody came in and

00:55:03.260 they were having stable angina, so they said, God, you know, doc, I'm walking up the hill and I get

00:55:07.200 this burning in my chest and I sit down and it goes away and I can't, now I'm just not able to walk

00:55:12.700 anywhere. So what can we do? Well, people realize that if you went in there, you blow up the balloon,

00:55:18.520 this angioplasty thing, but if you wrap a metal scaffold around, it's almost like a slinky around 0.95

00:55:23.360 the balloon. When you blow up that balloon, you leave that behind and that helps to keep the artery

00:55:28.180 from recoiling back down. And so that was more of a physical barrier to the artery coming back down.

00:55:33.540 So people found that if you just did the balloon, this is back in the 1990s, again, in people with

00:55:38.280 stable angina, a certain percentage of them, probably 40, 50, 60% of people would show up

00:55:43.840 with what was called restenosis. And restenosis just meant that the artery would narrow back down

00:55:49.180 again. Some of that was thought to be physical. Some of it was thought to be secondary to the injury

00:55:52.920 you cause when you blow up the balloon. So people thought, well, gosh, if you could leave this metal

00:55:57.600 scaffold behind, you can actually improve the likelihood that it's not going to restenose.

00:56:01.680 So stents completely cured the physical recoil problem, which I think, you know, again,

00:56:07.380 I don't know the exact number, but it was some non-zero number, but they left behind this other

00:56:12.820 problem, which was the injury problem. And so what people realized was that in the act of blowing up

00:56:17.200 the stent, you created a lot of injury on the vessel wall. And so the body's reaction to that was to

00:56:21.420 create scar tissue. And so then you'd get what now was called instant restenosis or ISR, instant

00:56:27.620 restenosis, which was basically the formation of a bunch of fibrous scar tissue inside the stent.

00:56:32.260 So before the era of what are now called drug eluding stents, this is back in the 1990s. If

00:56:37.980 you did a stent on somebody, again, stable angina, you'd see that there was some likelihood that they'd

00:56:43.720 come back within three to six months with recurrent symptoms. And you'd do a stress test. You'd see

00:56:49.340 that there was looked like there was ischemia. You'd come back in, you angiogram, you look at the

00:56:53.340 artery and there was narrowing again within the stent. And that was called instant restenosis. We went

00:56:58.180 through this entire period of five or seven years of trying to find ways to solve restenosis,

00:57:03.200 including radiating the inside of the vessel wall. So, but when I was a cardiology fellow,

00:57:07.140 we'd have the radiation oncologist come into the cath lab with us and we would put these

00:57:11.040 radioactive beads up like in one of these balloon catheters into somebody's artery and basically

00:57:16.620 try to almost kill the fibroblasts and other cells that were going to make the scar tissue.

00:57:21.840 And God knows what we were doing to each other and to the patients. But that went away

00:57:25.280 in the early two thousands with the advent of order called drug eluding stents. And we can talk

00:57:29.200 about those, but you probably already talked about them a bunch. Well, no, no, I was just going to

00:57:33.240 make the point, right? Which is we've sure talked a lot about my favorite drugs that end up in these

00:57:38.880 drug eluding stents. But why don't you take a moment to explain why a drug that inhibits mTOR

00:57:45.100 would find its way onto the coating of a metal stent? I can't remember the exact story. I think it

00:57:51.400 involved a guy who used to be at Columbia named Andy Marks. I think he actually figured this stuff

00:57:57.340 out, but we'll have to go back and do a deep dive on that. But basically these drugs, these

00:58:02.460 rapamycin and other drugs, these mTOR inhibitors were known to be, I think first, before even they

00:58:08.700 were modulators of the immune system, they were known to be anti-proliferative drugs that people

00:58:12.460 use to treat cancer. And so again, what I described in a simple way that I think about it is that it's a

00:58:16.980 proliferation problem. So these cells that are making, in response to an injury, a real injury,

00:58:21.560 they're making scar tissue. And so somebody realized, gosh, if you apply this drug, you can

00:58:28.760 block that proliferative process. You can block the smooth muscle cells and fibroblasts and other cells

00:58:33.240 that are making this scar tissue. And what they realized was that you could paint the outside of

00:58:39.700 a stent with this drug. And so it would basically be there locally and it wouldn't cause any systemic

00:58:44.960 toxicity or any other problems. And it would just act on the cells that you wanted it to.

00:58:50.320 So that mostly got rid of that problem that we used to call instant restenosis. For the most part,

00:58:56.740 now that problem is no longer. And there was a bit of a hiccup in 05, 06, 07, when the drug

00:59:04.920 eluting stents were briefly taken off the market. And my recollection is you had Medtronic still had a

00:59:11.800 bare metal stent that was basically getting all the market share while Boston Scientific and Abbott

00:59:17.880 were sitting their time out waiting for these drug eluting stents to come back. And I recall that

00:59:22.880 they even, at one point, even Boston Scientific and Abbott had the same stent, identical, but branded

00:59:27.800 under different names. I mean, it was such an interesting time in the cardiology world.

00:59:31.540 I don't remember all the details. I remember a lot of these reps from these companies were in the

00:59:35.500 cath lab and there was a big battle to get market share in these different stents. And there were

00:59:40.020 other things besides which drugs were on the stents. There were issues around deliverability and how

00:59:45.220 stiff they were and how long, you know, all these other things. And so it was a, you know, as a

00:59:49.440 general cardiology fellow and somebody who was interested in basic science, I wasn't paying

00:59:53.740 attention to the nitty gritty of kind of what was going on in this war for market share, but there

00:59:59.240 definitely was a lot going on. The biggest problem with these drug eluting stents was, so if you back up

01:00:03.460 to the early 1990s, when stents first started getting used in coronaries, one of the biggest problems that

01:00:09.120 happened was that you'd put the stent in there and then within a period of a few weeks, a non-trivial

01:00:15.220 amount of people would come back with effectively was a STEMI. The stent would clot off. That was

01:00:20.080 happening and that was bad. So there were series of trials where people tried aspirin. I believe

01:00:25.060 they tried warfarin. They tried everything and they actually couldn't.

01:00:27.880 So aspirin and warfarin are both blood thinners through different mechanisms.

01:00:31.200 That's right. So aspirin acts on platelets, warfarin blocks that coagulation cascade,

01:00:35.520 you know, downstream tissue factor. But people couldn't use these stents for a while. And then

01:00:40.420 in 1996, and I know this because I, as a medical student, had the ridiculously good fortune of being

01:00:45.280 the first author on a paper that was published in the New England Journal of Medicine. And that same

01:00:49.640 issue, so our paper is forgotten mostly except for me. But in that same issue, and I looked at it the

01:00:56.360 other day, was a trial that demonstrated that you could do stenting of coronary arteries safely

01:01:01.940 with the addition of this drug that's no longer used because it had toxicity called

01:01:05.980 ticlopidine, ticlid. And so ticlid was the first generation of these, what we call 0.91

01:01:09.980 thianopyridines. Thianopyridines act to block, at the time we didn't know this, but they block

01:01:14.060 a receptor on the surface of a platelet that's called the ADP receptor. And it basically renders

01:01:19.120 the platelet a little bit less sticky or less prone to be activated and become sticky.

01:01:24.280 And what these guys demonstrated, which was transformational for interventional

01:01:28.200 cardiology, is if you gave people this drug, ticlid, at the time they got their stent,

01:01:32.800 the chance of the stent clotting off went to almost zero. And eventually that drug had some

01:01:37.980 toxicity and was replaced with the drug that's used in most cases still today called clopidogrel,

01:01:43.880 which is acting on the same receptor. It's the same biology.

01:01:47.320 And patients might know that drug by a different name.

01:01:50.140 Plavix.

01:01:50.560 Yeah.

01:01:50.720 Right.

01:01:51.360 So today when a patient gets their stent, they also get a, what, a year supply of Plavix?

01:01:56.260 How long do we keep it on?

01:01:57.440 Yeah. So where I was going with that was that there was this thing called subacute stent

01:02:01.460 thrombosis. So thrombosis is the word for blood clotting. Subacute means within a few weeks.

01:02:07.140 So back in the day, if somebody stopped taking their Plavix or didn't get their Plavix, there

01:02:12.700 was a likelihood that they would get this condition called subacute stent thrombosis. Again, for all

01:02:16.720 intents and purposes, it's like a STEMI because it's a complete stoppage of blood flow and it's

01:02:21.480 critical and people can die.

01:02:22.660 Which is kind of ironic, right? You can show up with a stable problem that's chronic and through

01:02:28.540 the intervention end up having an acute fatal complication. I mean, the stakes are so high.

01:02:34.040 Huge. And people did, and it was bad. And, you know, there were lots of questions about why and

01:02:39.160 all that was happening. That problem mostly got solved for the, again, 95 plus percent of people.

01:02:45.040 There was some discussion about Plavix resistance and other things, but mostly that problem got solved

01:02:49.520 with Plavix. And people realized if you take Plavix for a month after a bare metal stent,

01:02:53.360 you're good to go. So then people fast forward and you were in the drug-eluting stent era and we've

01:02:58.500 solved the restenosis problem, but people are showing up now. They've taken their Plavix for a month,

01:03:03.400 but now they're showing up at three months, six months, nine months with not subacute, but whatever

01:03:09.040 the after subacute is, stent thrombosis, late stent thrombosis. And having, again, the same

01:03:14.440 life-threatening STEMI-like experience. And what people realized was in the process of inhibiting

01:03:21.040 these cells, these smooth muscle cells and other cells that make the scar tissue, we're also

01:03:25.340 inhibiting the endothelial cells. So one of the things that happens if you take a patient who has

01:03:29.880 a stent and that patient, say, dies in a car accident, you look under the microscope pathologically,

01:03:35.120 you'll see that that stent is now covered with endothelial cells after about three weeks.

01:03:39.540 And it's the endothelial cells that prevent the blood clotting. So the stent struts themselves

01:03:45.080 are probably pro-thrombotic. So they probably trigger some clotting. So what we realized was

01:03:49.820 that your own body would basically wall off the stent. Which is exactly what you want.

01:03:53.940 In a good way. It's to reproduce the endothelium that is this beautiful, you know, I try to explain

01:03:58.800 to patients that, you know, atherosclerosis is complicated. And the reason that there's no one

01:04:03.660 single causal factor is you have to have this sort of perfect storm of endothelial injury.

01:04:08.200 And that can be anything from smoking cigarettes will do that to high blood pressure will do that.

01:04:13.000 You have to have a lipoprotein that can actually carry cholesterol through that injury. So you have

01:04:19.240 that injury, you can carry it through. That has to get oxidized and create an immune response

01:04:22.900 that then has to lead to the formation of this plaque, which for reasons that aren't entirely clear,

01:04:27.520 some rupture, some don't, and kick off the epoxy problem. So what you just described is

01:04:32.800 the body will reproduce that beautiful endothelium around the metal stent again. And if that's what

01:04:39.280 happens, you end up in the good camp. And if instead of that happening, you kick off this

01:04:45.240 inflammatory cascade again, you can get into a very bad situation very quickly.

01:04:49.600 Basically, the way to think about it is that most people with a bare metal stent

01:04:53.240 will re-endothelialize the inside of the stent within, you know, like three weeks, four weeks.

01:04:57.540 So the standard of care in 2002, 2003, but before drug-eluting stents was take Plavix for one

01:05:02.300 month and you're done. And people did great. And then drug-eluting stents come along. They solve

01:05:06.940 this one problem, but they create a new problem. The problem was that in addition to inhibiting the

01:05:11.180 proliferation of these smooth muscle cells and fibroblasts-

01:05:13.460 Inhibit the endothelium.

01:05:14.280 Exactly. So people were end up coming back with this stent thrombosis problem now, you know,

01:05:19.160 months later, then it became unclear about what the correct amount of time to take your Plavix or other

01:05:25.780 drug was after a stent. And I think now there's some great work being done by people. I think for

01:05:31.300 the most part with these newest generation drug-eluting stents that have the right mixture

01:05:36.320 and cocktail of drug on them, and they're really easy to deliver and cause less injury, it's six

01:05:41.900 months. If something's complicated, if the patient has diabetes or other risk factors, then you may

01:05:47.920 extend the time. But again, what you're doing is you're buying bleeding risk by inhibiting this risk

01:05:54.020 of stent thrombosis. Obviously, stent thrombosis is the worst. You can't have that. So when I have

01:05:58.920 a patient who leaves the hospital and they say, what do I take? I say, listen, if you're going to

01:06:03.880 an island and you can only bring one thing with you for the first, you know, whatever it is, few

01:06:07.680 months, all that matters is your Plavix. You've got to take it every day.

01:06:10.860 Right. At this point, whether you remember to take your statin or your blood pressure drug,

01:06:15.680 those are important. But yeah, that's a great way to put it. If the ship is sinking,

01:06:20.600 the highest priority is the lifeboat. This is your lifeboat.

01:06:25.100 It didn't get communicated clearly enough, I think, for a while. So when people would show up

01:06:28.420 in the hospital with stent thrombosis, it was often because they hadn't been told

01:06:31.880 appropriately, strongly enough that they had to take Plavix.

01:06:35.480 So you've done a great job of explaining the unambiguous part of interventional cardiology,

01:06:41.860 which is the case where I think most people agree interventional cardiology through the stent

01:06:48.540 has provided a survival benefit. There are a whole bunch of other areas where that is quite gray.

01:06:54.280 You and I have had some really interesting discussions about that going through some of

01:06:58.440 the really important trials in the past decade that have tried to ask a question. So let's begin

01:07:03.800 with the case that you described a while ago, which is patient comes to see you in clinic says,

01:07:09.540 you know, Dr. Weiss, God, it's just every time I get on that treadmill, I just get a tightness in

01:07:13.760 my chest. And the other day I was walking up the street and it was really steep and blah, blah, blah,

01:07:17.980 blah, blah. So they've got stable angina. Let's assume you decide it's worth doing an angiogram

01:07:24.200 in that person. You do. And you see, lo and behold, there is a 60 or 70% occlusion smack in the middle

01:07:30.840 of their left anterior descending artery that runs right down the front of their left ventricle.

01:07:35.940 What should you do? Maybe another way to say it is what would people have normally said and what

01:07:40.260 have the data now suggested? I mean, this is one of the most controversial areas, I think,

01:07:45.000 and least well-understood areas of medicine. I'm not even going to say cardiology. So in the 1980s,

01:07:50.720 that patient gets a prescription for nitroglycerin and beta blocker. So beta blocker

01:07:54.520 helped reduce the demand. So basically it's like putting a governor on a car or golf cart. So you

01:07:59.440 can't rev the engine as high. So you're going to be less likely to get symptoms. And nitroglycerin,

01:08:04.020 as we described, helps relax the blood vessels and maybe also decreases demand by decreasing blood

01:08:09.860 pressure. So it's basically a couple of not great tools in the 1980s. And then people would go to a

01:08:16.500 point where they couldn't walk anymore. And eventually they'd get bypassed because there

01:08:19.880 was no other option for what we now call revascularization. Eventually when stents became

01:08:25.280 viable in the mid 1990s, people realized, well, gosh, that person, if you open that artery and put

01:08:30.800 a stent in there, you can give somebody a really durable response and they won't have to take any of

01:08:34.600 these medicines. And just to be clear at this point in time, we only had one sort of plumbing

01:08:41.480 solution that had been demonstrated to impact survival. And that, if I'm correct, it was Lima

01:08:47.200 to LAD cabbage. So I'll explain what that is. So everything, as you can tell in cardiology is an

01:08:52.300 acronym. So Lima stands for left internal mammary artery. So when you open up the sternum, you have

01:09:00.120 these arteries on the inside of the sternum and the left and right version of these turn out to be the

01:09:07.660 most important conduits that you can use to bypass because they're already attached on one end and

01:09:15.700 you're, they're beautiful blood sources. And so you just attach it, what we call distally. So past the

01:09:20.800 lesion and, and by the way, if my, my memory is correct, it was only Lima to LAD. It wasn't Lima

01:09:28.080 to Cirque. It wasn't Rima to right main. There was nothing else. They did sham surgeries. If I recall 0.93

01:09:34.640 to demonstrate that. I don't know if they did sham surgeries back then, but you're absolutely right.

01:09:39.380 It was only Lima to LAD and it was only a subset, right? So I think it was, you had to have

01:09:42.900 multivessel disease, left ventricular dysfunction, diabetes. I mean, there were a subset of people.

01:09:47.560 Not everybody got a mortality benefit. I'd have to go back and look at the,

01:09:50.980 maybe those weren't the shams, but I know there were some sham surgeries that involved sternotomies, 0.94

01:09:55.180 which when you think about that, I mean, imagine signing that consent form. Golly.

01:09:59.740 So in defense of the cardiology community in the mid nineties, it wasn't an unreasonable hypothesis

01:10:05.500 that, Hey, if taking the Lima and attaching it to the LAD, effectively bypassing this thing works,

01:10:12.320 at least in a subset of patients, shouldn't we be able to rotor rooter this thing and open it up

01:10:17.240 and get the same benefit? Right. Right. And I think the important thing to remember is to define

01:10:22.220 the benefit you're interested in. So we need to talk about the two things that matter. And when a

01:10:27.180 patient comes to see me in the office or you in the office, we talk, we often end up talking about

01:10:30.220 two things. One is how do I feel? So presence or absence of symptoms. And the second is, is this

01:10:35.500 going to impact how long I live mortality? And those are the two variables we think about every day in

01:10:41.140 cardiology. And as you mentioned, there was evidence that this one procedure or a subset of these

01:10:46.820 procedures could confer a mortality benefit. But most of the time when we're talking about

01:10:50.800 revascularizing or treating stable angina, we're talking about treating symptoms that remains true

01:10:56.340 today. And people sometimes poo poo that, but I think it's important, at least for me, I like to

01:11:00.720 dispel that because there's certainly a, I find a heavy dose of people in the peanut gallery who think

01:11:07.340 if it doesn't impact all cause mortality, it's never worth talking about. And I suspect these people

01:11:12.620 have never taken care of patients. Of course not. No. I mean, look, if you can't walk up your

01:11:16.760 stairs, you can't garden, you can't hike with your wife, you can't go skiing with your children. I

01:11:22.260 mean, there, there are a lot of things that were- Can't have sex. Yeah. Yeah. Yes. And all that

01:11:26.340 stuff I think is incredibly important and it, it shouldn't just because you're not conferring a

01:11:31.040 mortality benefit, all cause mortality benefit doesn't mean that there's not benefit. So I think

01:11:35.160 that the advent of the stent era was, was remarkable. It happened to coincide with the advent of the staten

01:11:40.520 era, right? So the first stat and lowest stat was approved in 1988, I believe. So remember I had said

01:11:45.940 that we were treating people with angina with basically beta blockers and nitrates. And so

01:11:49.360 then statins got added to the mix. They weren't initially added for treatment of stable disease.

01:11:54.420 They were there for, they were initially tested in people who were having unstable disease, but it

01:11:59.340 quickly became obvious that they were going to become an important mainstay of treatment of people

01:12:03.380 with stable coronary disease as well. So this idea that even if there's an intervention that doesn't

01:12:07.580 extend life by one day, if it improves the quality of life, it should still be on the table.

01:12:11.900 So walk us through a little bit of the trial architecture and how it navigated how we ought

01:12:19.300 to treat these patients with stable angina, or you can even define it as the, if you want to do it

01:12:24.560 through the ACS language, feel free as well. So let's back up and just make sure we reemphasize,

01:12:29.260 cause I don't think this point cannot be emphasized too many times. So if you show up in the hospital

01:12:32.620 or in the emergency room or anywhere in the world with a ST elevation MI, a STEMI, in other words,

01:12:38.580 that ST segment is increased above baseline and you're having a heart attack.

01:12:42.480 Yeah. It's do not pass go, do not collect $200. You're going straight to the calf.

01:12:46.440 That's right. If you show up and you have a non STEMI, so chest pain, let's say you have

01:12:51.980 positive biomarker. So now modern day we use another biomarker called troponin, which is just a

01:12:56.660 component of the architecture that allows muscles to contract. And there's an isoforma, a flavor of it

01:13:02.940 that's specific to cardiac muscle. So if you show up with some of that in your blood, it signals that

01:13:08.220 your heart has suffered cell death, muscles have. And so that's part of the diagnosis. So you show up

01:13:13.960 with chest pain, positive troponin and an EKG change that is not ST elevation, anything other

01:13:19.680 than ST elevation could be normal. That puts you into a different category. And mostly in the United

01:13:25.480 States, the management of non, depending on a couple of factors, but most people are managed

01:13:30.700 either what's called medically. So with a bunch of medicines or they go down a path of interventional

01:13:36.360 cardiology. So now let me ask you a question, Ethan. We didn't distinguish this, but it's worth

01:13:40.480 adding in the STEMI patient. You don't care if they're hemodynamically stable or not. In other

01:13:47.460 words, if their blood pressure is stable, you know, they're having chest pain, but they're not

01:13:52.440 in extreme distress. They're still going to the cath lab in the non STEMI patient has chest pain.

01:13:59.680 Patient has the enzymes. EKG looks normal, but presumably there are still cases where those patients are

01:14:05.560 not hemodynamically stable. So if you have what we would call complicated non STEMI, so

01:14:10.500 hemodynamic instability, I mean, your blood pressure is very low or you have heart failure,

01:14:15.000 your fluid in your lungs, something else that makes it complicated, or you can't treat people,

01:14:19.560 can't make their chest pain go away with medicines, something that makes it complicated or ongoing

01:14:23.480 concerning EKG changes, or just you don't feel right. That patient also can go to the cath lab.

01:14:29.100 It isn't an emergency unless there's hemodynamic instability, but oftentimes people will go

01:14:34.260 along the same timeline. It may not be within 60 minutes, but it'll be quickly. Again, depending

01:14:39.980 on how complicated. If you have uncomplicated non STEMI, so the blood pressure is fine, you don't

01:14:44.240 have any evidence of congestive heart failure. There's nothing else that makes this look scary.

01:14:48.660 Those patients can be managed either non-invasively with medicines or invasively. And we can talk a little

01:14:53.680 bit about how that happens. There was a period of time, I think, in this country where most of those

01:14:58.560 people went into the interventional arm where they went and they went to the cath lab and they would

01:15:02.440 get a stent. And I think in the past few years, there have been questions about the sort of value

01:15:07.920 of that and there's debate about that. But just to be clear, there's one area of medicine where there's

01:15:12.240 no debate, which is there's a more clear mortality benefit by opening up the artery. Again, if you're

01:15:17.640 in Uganda and you don't have a cath lab, then TPA is the best you can do. But opening the artery saves

01:15:22.940 lives. In the non-STEMI situation, if you're hemodynamically unstable, again, same thing.

01:15:28.560 Opening the artery saves lives. If you're hemodynamically stable, the things start to

01:15:32.620 become a little bit more gray. And then stable angina, which is the last thing we'll come back

01:15:36.800 to, there's a huge amount of gray area. Again, about mortality, not symptoms.

01:15:41.940 So let's talk about Orbita. What did that trial look at? What question did it try to answer?

01:15:46.800 This trial was published in November of 2017, I believe. And it was done by a group in London

01:15:52.820 who I had never heard of before this trial came out. I guess Daryl Francis was the PI. And I think

01:15:58.660 he was well-known in the interventional cardiology community, but not to those of us who are not in

01:16:02.420 that world. This trial shook the world. Actually, we should probably talk about CURGE before we talk

01:16:08.120 about it.

01:16:08.380 Let's do that. Yeah. Thank you. Let's go. Because I alluded to CURGE and then we got off topic.

01:16:12.020 Tell me about CURGE.

01:16:12.820 So CURGE was published in 2007.

01:16:14.660 And if anybody asks me what CURGE stands for, I don't remember.

01:16:17.620 I don't remember either.

01:16:18.480 Yeah. But it's an acronym that is put together to explain what the trial was about.

01:16:23.000 And I also can't remember who the PI was, but that study. So again, we're talking at this point

01:16:27.820 back about stable angina. So people who are not showing up in the hospital with new symptoms or

01:16:33.660 new EKG changes, this is a totally different animal. This is somebody who's been out living in 0.59

01:16:37.780 their home and they have symptoms that have been going on for some period of time. So CURGE was

01:16:43.380 designed to ask the question and answer the question of, is there, in addition to a symptom

01:16:49.440 relief that you'd get from putting a stent in an artery of somebody who's got a blocked artery and

01:16:53.140 has stable angina?

01:16:54.100 By the way, I'm just going to interject. So I'm going to make a plug for one of my favorite

01:16:57.040 apps, which is incredibly dorky. It's called Trials. I don't know if you've seen this app,

01:17:02.580 but let me show it to you and we'll make sure we link to it. So the, I'm sorry, the app is called

01:17:05.940 Journal Club. It's either free or very cheap. So if you go into Journal Club, you can search

01:17:11.480 every single clinical trial in the history of mankind. So I just went COURAGE and it pulled

01:17:17.900 it up and it's, it gives you just a treatise on the topic, right? It gives you the bottom line

01:17:22.260 and then it tells you what it stands for. Okay. So major points. So it's called Clinical Outcomes

01:17:26.040 Utilizing Revasculation and Aggressive Drug Evaluation, parentheses, COURAGE. And then it goes

01:17:33.140 through and explains everything about it. So if you're listening to this and you're a physician

01:17:36.600 and you don't have this app, get it. And truthfully, even if you just have a tincture of interest in

01:17:43.560 clinical trials, this is a great app to have. I must look at this like once every two or three days

01:17:49.620 and you can screen and search by like, sometimes if I'm just sitting around and I've got 20 minutes

01:17:55.420 to kill and I'm kind of bored and it's, you know, I'm in the, I'm on the runway and it's taken a while

01:17:59.480 to take off. I'll just open it up and search by subjects. Like, I wonder what's going on in oncology

01:18:03.780 in the last two months. What new trials have come out? So anyway, Journal Club is awesome.

01:18:09.060 Well, now you can read it and see how close I am. So I'll try to, I will be intentionally less

01:18:14.160 detailed about how I described these trials since everyone's going to be reading along.

01:18:18.280 So anyway, this trial was designed with one purpose. And the purpose of this was to compare

01:18:21.640 maximum medical therapy to stenting in patients with stable coronary disease. And the idea was,

01:18:26.820 is there a benefit in terms of reducing the risk of heart attack or some other heart outcome like death

01:18:32.300 in patients when you randomly assign them in a non-blinded manner to be treated with maximum

01:18:38.080 medicines or stents? Non-blinded meaning they did not do sham procedures in the people who randomized

01:18:44.020 to no stent. That's right. So obviously you as a patient or you as a doctor know if you're going

01:18:49.340 to go get a stent or not. So you, you were told you've been assigned to this category. You're going

01:18:53.960 to get medical therapy and then they got medicines. And I, this is something I'll have to go back and look

01:18:58.500 up. But I believe the people randomized to the interventional arm also got aggressive medical

01:19:02.400 therapy. I don't think there were huge differences, but that may be one of the areas where the,

01:19:06.420 there was debate, but the net of this whole thing was that this trial was organized and run by

01:19:11.400 interventional cardiologists. And the purpose of it was to demonstrate that there's a benefit in

01:19:15.880 heart outcomes by stenting that this, we were going to prove finally that opening the arteries

01:19:20.660 did more than just relief symptoms. And it was negative, completely negative. So there was no

01:19:25.840 difference between the two arms. I can't remember the details because it's been over 10 years,

01:19:30.420 but there may have. I remember that because like I said, I just saw that it was 2007. I was going to

01:19:35.320 guess 2006, but I kind of remember where I was when that study came out and boy, that was like a big hit

01:19:43.600 in an industry where it was pretty much the wild West at that point in terms of the number of stents

01:19:49.920 people were getting. I mean, we used to joke about it, kind of teasing these guys, which is like,

01:19:54.720 God, if somebody looks at you wrong in the parking lot, you're going to put a stent in them.

01:19:58.720 Well, it was, and this was the era again, where there were people like there was this hospital up

01:20:02.520 North. I think it was like a tenant hospital where there was this guy who was stent. You'd show up

01:20:06.520 there with like too much farting or something. And then he'd do a calcium scan on you and you'd see

01:20:11.080 like a, you know, calcium score of 20. And he'd say, we got to take you to the cath lab. And then he'd

01:20:15.420 put stents in you. And there was really no way to argue against that. Even in asymptomatic people, 1.00

01:20:20.580 it was just a crazy time. So this, this definitely was a big result in cardiology,

01:20:24.500 particularly in interventional cardiology. It was not met without criticism, as you can imagine.

01:20:29.360 And I can't remember all the details, but there were definitely some big questions raised,

01:20:33.020 but the net of it was, Hey, look, the other way to look at this is that we've been doing trials

01:20:37.300 of stenting for a long time now. And no one outside of the STEMI situation has demonstrated any benefit

01:20:45.460 in hard outcomes that to date. So, you know, you could say the burden of proof, the onus is on the

01:20:51.060 interventional cardiology community to prove it. So they tried and they failed. So I think

01:20:54.740 that was a big practice changer. I think what that told you was, look, again, back to what we

01:21:00.060 were talking about earlier, this is a symptom management problem. We're not doing this to

01:21:03.940 extend life. It doesn't mean that it's not important. It doesn't mean that you still don't

01:21:07.640 use it as a tool, but it may be, it shouldn't be the primary tool. Maybe the primary tool should be,

01:21:13.340 Hey, look, let's try and use these medicines, which are really effective and we'll optimize them as best

01:21:17.760 we can. And if that doesn't work, then we'll go down the path of trying to open the artery. And so

01:21:21.980 it really did change. I think the way we managed people again, mostly outpatients, mostly stable

01:21:26.680 angina patients. So then what was the impetus for Orbita 10 years later? Well, I would have to ask

01:21:34.380 Daryl and I hope someday you could ask him cause he's a phenomenally entertaining guy. Yeah. You got

01:21:39.020 me onto him and his work and I follow him on Twitter now and I think he's in the UK and I can't,

01:21:44.200 I can't wait to hopefully meet him at some point. He made this like tremendous appearance and splash

01:21:49.460 on Twitter a year and a half ago and like set the world on fire cause he was fearless and obnoxious

01:21:54.860 and just everything you'd love about him. And, uh, and I guess he pissed a lot of people off and

01:21:59.140 somebody basically pulled his plug. So he's been really quiet, if not completely absent from Twitter

01:22:03.780 for the past, I want to say six months. But anyway, he was the PI on the study Orbita and the impetus

01:22:09.360 there was, okay, well, what courage kind of established that there's not going to be a benefit in terms of

01:22:12.860 heart outcome. So we're really, we've sort of settled on this idea that stents are there and

01:22:16.980 they're useful for treating symptoms in patients who failed medical therapy. But Orbita was our

01:22:23.500 stents actually even doing what they're supposed to do in terms of reducing symptoms. So Orbita,

01:22:27.880 you mentioned sham trials. This was the first sham trial that I, interventional cardiology that I knew

01:22:32.500 of. So Orbita stands for objective, randomized, blinded investigation with optimal medical therapy

01:22:39.580 of angioplasty in stable angina. And one of the beauties of naming clinical trials is you can use

01:22:46.900 a sentence with as many words in it as you want. As long as you capitalize the beginning of the words

01:22:52.280 that you want in your acronym, you get it. So if you just heard me read that and said,

01:22:56.600 how the hell does that work out to Orbita? You only capitalize the objective, randomized, blinded

01:23:02.640 investigation therapy, angioplasty. Yeah. Well, that app would have been fun as a medical student

01:23:07.760 because you used to get pimped on kind of what were the key trials and stuff. So anyway,

01:23:12.000 the key differentiator of this trial is blinded, which meant there were sham procedures done. So

01:23:16.460 what they did was they took people who had demonstrated, documented angina and lesions.

01:23:22.160 So they all had catheterizations. Whereas I think in Courage, they were randomized after a stress test.

01:23:27.620 I don't think everybody had an angiogram before they got randomized. In this case, they got randomized

01:23:31.500 after the angiogram showing that they had a blockage. They could not have three vessel disease or

01:23:37.240 left main disease. I think there were a few other exclusions, but these were people who could have

01:23:41.080 one, two advanced disease and they could be tight lesions. They could be 95, 98% lesions.

01:23:47.860 Then they were randomly assigned to get either optimized medical therapy or a stent. And that

01:23:53.560 was done blind to the patient and blind to the, not to obviously the interventional cardiologist,

01:23:57.820 but to the treating, referring physician. So if I sent you in to be randomized in this trial,

01:24:03.520 you and I would not know what you got because they would take you to the cath lab. They'd put

01:24:08.060 a catheter in your leg. They'd wind a catheter up into your corner or ostium, and they would pretend

01:24:13.260 to blow up a balloon and you wouldn't know what you got. And it was a small trial, which was one of

01:24:18.360 the major criticisms. I think it was such a small trial and I'm not a trialist, so I don't get into

01:24:23.760 the nitty gritty on kind of what makes trials, you know, more or less robust.

01:24:28.500 Well, I mean, from a size perspective, the issue comes down to power. I think this is the biggest

01:24:33.320 challenge of trial design. People talk a lot about significance, which is what the P value tells you,

01:24:39.520 but people don't talk a lot about power, which is beta. And the idea here is you have to have some

01:24:45.740 sense a priori of how much of a difference you will see to design your trial because you have to

01:24:52.360 know how many people to put in it. Because if the trial comes out and there is a significant

01:24:57.700 significant difference, a statistically significant difference, which is usually

01:25:00.860 defined as a P value less than 0.05, meaning there is a less than 5% chance that the difference you

01:25:06.580 have observed is due to chance, then power becomes irrelevant. It's just that power becomes so

01:25:12.440 important when the opposite happens, when you don't see a difference. And it's your ability to say,

01:25:17.660 we don't see a difference and we believe that a difference is not there versus we don't see a

01:25:23.360 difference, but we didn't have enough people to look at it. So for example, if I was going to

01:25:27.660 design a trial with me and Ethan, just as the two of us, it's quite likely I would never find a

01:25:33.460 statistically significant distinction. But that doesn't mean that the metric I'm looking at doesn't

01:25:38.500 exist. It's just with two people in a trial, the probability that I'm going to find it is incredibly

01:25:44.000 low unless the difference is egregious. So I've always found this to be the biggest challenge of

01:25:49.920 clinical trials is that you have to have some sense of your expected outcome to select beta to power

01:25:57.660 the study appropriately. And this was the, obviously the, the core place that people argued because it

01:26:02.860 was a small trial and people said it's an underpowered study and there's, you couldn't see a

01:26:07.060 difference. Study was interesting because they, they both asked people sort of, how do you feel in

01:26:11.420 terms of your angina? So how many, every one of these patients had some frequency of,

01:26:15.080 of anginal episodes. They'd get chest pain if they walked or did something, but then they did some

01:26:20.160 objective things too. So they did stress testing. I believe they, they did like the obedient stress

01:26:25.000 test. So basically they, it's a way of looking to see objectively was there more or less ischemia.

01:26:30.620 And they also did treadmill tests. So they looked at their exercise capacity and how long they could

01:26:36.840 exercise before they got symptoms. And I don't want to get misspeak and piss off all my interventional 1.00

01:26:43.200 friends. But for the most part, the thing was pretty negative. And I think there might've been

01:26:47.380 a slight difference in the treadmill time. There was something very slight, but for the most part,

01:26:51.480 it was a negative study. And so the result, which ended up splashed all over the New York times,

01:26:55.800 the wall street journal and every other, you know, news organization you could think of the,

01:27:00.400 the conclusion that was sort of reported to the general public was since not only do they not

01:27:06.520 reduce your risk of having a heart attack or dying, they don't even reduce your symptoms. And so

01:27:11.500 it became a sort of firestorm of conversation.

01:27:15.420 And this was not long after former president George W. Bush had a stent placed after what I

01:27:21.720 don't believe was a STEMI.

01:27:23.460 That's right. He had stable angina, same thing. And this goes back to the old days of when people 0.93

01:27:27.800 would do routine sort of surveillance stress testing and people to look for these 70% blockages,

01:27:33.700 and then they'd go and stent them, right? This was the way things operated until relatively recently.

01:27:38.800 So what we learned from that trial was, look, this may have been underpowered, but at the very

01:27:44.780 least, it tells you again and again and again, that medical therapy is really good. So modern

01:27:48.980 medical optimized therapy is great. And so the way I, what I took home from this trial and what I

01:27:55.500 convey to my patients today, along with the important caveat that what I convey to you today

01:28:00.720 is very likely to be a lot different from what I convey to you tomorrow. But I say what I convey to you

01:28:05.880 today is that while stents may or may not be useful in some circumstances at reducing your

01:28:12.300 symptoms, if you have bad symptoms, it's very unlikely that in stable angina, they're going to

01:28:15.980 do anything else. But either way, it's our job to try to manage you as best we can with medicines

01:28:21.980 before we go down that path, because there are going to be consequences of putting a stent in you.

01:28:26.280 And we've all seen these patients who end up with 20 and 30 stents, you know, tons and tons of

01:28:30.240 stents. And again, taking a stent is not just an isolated experience. It means going on one of

01:28:36.180 these powerful blood thinning medicines like Plavix or a more powerful one. And it means a lot of other

01:28:42.180 things that we do and don't understand. So what I say is, look, our job, our obligation is to try

01:28:47.180 as best we can to use medicines to optimize you. If we fail at that, we're going to keep that stenting

01:28:52.760 procedure in our back pocket. If you come back to me in three months and you say either one of two

01:28:57.260 things, either these medicines are not working and I'm still having a lot of chest pain, then we go

01:29:02.420 down, then we can say, oh, let's go look at opening the artery with a stent. Or if you say to me,

01:29:06.340 these medicines are making me feel miserable. Like I just, I don't like the way I feel. I want to take

01:29:10.640 fewer medicines. I also believe it's reasonable at that point to say, all right, let's go see what we

01:29:15.860 can do with a stent. So I have not stopped referring people to interventional cardiology. In fact, I probably

01:29:21.440 haven't changed dramatically how many people get sent. Courage probably did change it a little bit, but

01:29:26.020 for the most part, I'm sending the same number of people. I'm sending the same people

01:29:29.560 to the cath lab as I did before Orbita. Yeah. It's probably just the composition of

01:29:35.100 those patients is probably migrating constantly. And that's, I think some people, you know, you said

01:29:39.020 something sort of interesting that I think for some people is frustrating to hear, which is what do you

01:29:44.260 mean what you're telling me today could be different tomorrow? But that's unfortunately the nature of

01:29:48.440 science. I, one of the things I like to say, and I wish I knew who said it first, because I, I feel like

01:29:53.640 I'm plagiarizing it though. I don't actually recall is that all facts have a half-life. And in some ways

01:29:59.760 that's the single most exciting thing about science. I mean, it would be kind of boring if

01:30:04.160 all of the natural universe and the, and its laws were known. And this was just a, a game of memorizing

01:30:11.180 all those facts. I mean, I just don't think science would be nearly as interesting as the fact that

01:30:15.140 we're constantly in the dark trying to refine our knowledge.

01:30:19.140 You can say that on a micro level and a macro level. I mean, I always tell people that work

01:30:23.480 with me in our, in my lab, in our lab that, you know, if you know the answer to the experiment

01:30:27.660 before you do them, we don't need to, we don't need to experiment. Just stop. We can write it up,

01:30:31.980 right? What are we going to learn? And the other thing is that the results of these experiments,

01:30:35.560 whether they're experiments in a lab with cells and tissues or animals, or whether they're clinical

01:30:39.540 experiments that are done on lots of human beings, there is some likelihood that that result is

01:30:44.360 quote unquote wrong. Even a P value of 0.05 means that there's a 95% chance that it didn't happen

01:30:50.800 by chance, but there's a 5% chance that it, that result did happen by chance. So there is a constant

01:30:56.000 refining of what is truth or what is as close as what we can call to truth. And so I think we have

01:31:01.380 to stay humble. I love it when I'm wrong. Then there've been a few examples this year where I've

01:31:05.720 been wrong, brutally wrong. And it brings me joy. Like I, it's a great, I love being, I love it when

01:31:12.020 something defies my expectations. I don't know that I can say I enjoyed as much as you do, but

01:31:16.320 if I just look at constantly being challenged, I mean, you know, at the time of this recording,

01:31:21.020 there is just such a gauntlet being thrown down to challenge the dogma around utility of vitamin D

01:31:27.040 and sunscreen. That's something that I've been spending a lot of time reading about in the past

01:31:30.720 two months. And I have to tell you, I don't know that I would say I love the fact that I probably know

01:31:37.080 nothing about this now based on all of this new information that is emerging, but it's certainly

01:31:41.960 exciting. I mean, it is so great to say, wow, I get to go back and learn this whole thing all over

01:31:48.180 again. And you sort of have to hope that there hasn't been damage done in a previous paradigm

01:31:52.640 that's being turned over. You know, you brought up your lab. So I want to talk about that, but I'm

01:31:56.780 not ready to leave this topic yet because I want you to give us a quick or reasonably quick primer

01:32:03.200 on other things that tend to confuse patients, such as calcium scores versus CT angiograms.

01:32:09.740 And I even want to touch on heart flow in a minute because that comes back to it. So

01:32:13.300 I think the listeners know what a calcium score is in a CT angiogram is, but so just give this the

01:32:17.640 quickest sense of that. Cause I'm, what I'm much more interested in is what did the results tell us

01:32:22.220 as a cardiologist practicing in 2019, I struggle with the question of whether I'm going to help you or

01:32:28.060 hurt you that I feel this tremendous sense of uncertainty about whether I should be as aggressive

01:32:34.420 as I can picking up every rock and looking under everything and, you know, trying to optimize

01:32:40.300 to the best of my extent, my ability versus whether that may be the best thing I can do is leave you

01:32:46.540 alone. And you've probably seen examples too, where, I mean, I remember again, as a cardiology

01:32:51.580 fellow, maybe even as a resident where, you know, somebody would come in from an outside hospital, 0.99

01:32:54.880 sick as shit, just absolutely on death's door. And all we did was just turn off everything 0.99

01:32:59.360 and the patient got better because they were just over managed. And I think I struggle a little bit

01:33:04.680 with this sort of where I want to be in that spectrum and how aggressive I should be in looking

01:33:11.080 for say a cult coronary disease, which I think is a question you get a lot. And I get a lot, right?

01:33:15.760 A lot, the, one of the major reasons somebody comes to see me as a preventive cardiologist, they say,

01:33:19.600 am I going to die of a heart attack? And you know, what's my risk of dying from a heart attack? Or

01:33:23.840 my brother died of a heart attack at 44, what should I do? And I still don't have an answer

01:33:30.320 about how aggressive I should be in trying to understand it. But a lot of these tests we'll

01:33:35.340 talk about, I think, feed into that. And I think ultimately what we're missing, and I hope we can

01:33:40.360 eventually refine it and make it better, is a good way to predict disease risk in these chronic diseases,

01:33:47.160 these common chronic diseases like cardiovascular disease, metabolic disease, that we just don't now yet

01:33:52.060 have the tools to be able to say, you know, Peter, well, your risk is X, Y, or Z. And so therefore,

01:33:59.060 we should do this or this or this in terms of prevention, understanding that there's going to be

01:34:03.700 risk in each one of these things that we do. And there may be risk and even part of the process of

01:34:07.560 getting from here to here, point A to point B. So I'm glad you brought that up because it illustrates

01:34:12.080 the challenge that frankly can't be explained or rationalized or described on Twitter. Not to pick

01:34:19.800 on Twitter, but just to, so there's this idea which you've said, which is, I don't know sometimes how

01:34:24.940 aggressive to be or not to be. And what you're really saying is at the individual level with you

01:34:31.060 as my patient sitting in front of me, I don't know how aggressive to be or not to be. You're not asking

01:34:35.380 the question on average. And yet, what tool are you given to guide you? You are given a tool called

01:34:42.340 a clinical trial, which is by its very nature, all about averages. And so therein lies the mismatch of

01:34:50.740 what I've described as medicine 2.0. When I say described, meaning I'm writing about it in this

01:34:55.060 book I'm working on that hopefully I'll have finished by the time I'm alive or not alive. And

01:35:00.180 the idea is it's not to poo-poo clinical trials. It's just to acknowledge that clinical trials give

01:35:05.820 us great information on averages and the larger and more robust the trial, generally the more

01:35:11.500 heterogeneous the data. But you've asked a question that comes down to judgment. You know what it means

01:35:18.160 to be aggressive and you know what it means to be conservative and you have, you know what the

01:35:22.920 corners of that box look like. What you're asking is I could have two people in front of me that

01:35:28.300 superficially look similar, but actually one of them is probably going to have a better outcome

01:35:33.360 if I behave aggressively. And the other one might have a better outcome if I behave conservatively.

01:35:37.560 It's the challenge to figure out which one's which. If you're a hammer and everything's a nail,

01:35:43.040 even if you're acting as a hammer and nail in accordance with clinical trials, I suspect you are

01:35:48.380 still acting in a very blunt manner. A hundred percent. But I'm also talking about these areas

01:35:54.320 and I think prevention is a great example that are sort of outside the boundary of what's been

01:36:00.360 studied or is likely to be studied in the context of a clinical trial, right? I mean, there's not going

01:36:04.760 to be a clinical trial to answer a lot of the questions that I have about how to manage my

01:36:08.880 patients. And I feel the same way. I mean, prevention is not really amenable to this idea of

01:36:14.040 medicine 2.0, which is clinical trial, average outcome, short duration, simple intervention,

01:36:21.540 easy to measure outcome. It's the economic thing. I mean, you're a company and you want to get your

01:36:27.780 product to market, whether that product is a stent or a drug or whatever it is. And the best way to do

01:36:32.880 that economically is the shortest amount of time. And so you want to take the sickest people. So

01:36:36.280 these trials, I mean, I joke that like a prevention trial, all kinds of trials that I want to do would

01:36:40.920 take 50 or 60 years. How do you convince somebody I'm about to be 50? I wouldn't want to start a trial

01:36:46.140 that I'm not going to see the answer from, the result from. So it's unsettling to me. And again,

01:36:51.480 I think you just have to remain humble as I've tried to and hope that your patients,

01:36:55.980 your human patients have some patients that were going to be wrong. There are a litany of examples

01:37:00.660 like LP little a was something I didn't pay attention to until the past few years. Coronary

01:37:05.480 calcium scans. If somebody came to see me with a coronary calcium scan 10 years ago, I would say,

01:37:10.300 I wish I didn't have this information, but I never ordered one before seven to eight years ago.

01:37:15.340 So there are lots of examples of things that I didn't used to do that I've now incorporated into

01:37:18.660 my practice. And I'm doing so without that like safety belt of, of evidence basis that we're used

01:37:25.300 to, right? There's not going to be a orbital like trial to help me decide whether I should be

01:37:31.020 aggressive with lipid lowering in a 35 year old. That's not going to happen with primary prevention.

01:37:36.200 So we have a mutual patient in whom that's exactly the type of question that's being asked,

01:37:40.680 right? Yeah. And there's a term and I, I know all these cute little terms and I never know who to

01:37:45.140 attribute them to, but we talk about evidence-based medicine versus evidence informed medicine.

01:37:50.280 And to me, the latter just makes much more sense because these decisions that you have to make

01:37:55.820 virtually every day. And I feel like I'm in the same situation, virtually nothing that I do.

01:38:02.640 Can I point to the orbita or courage equivalent? I mean, it just doesn't exist. And certainly not,

01:38:09.200 if you really wanted to scrutinize it, every single thing is a variation on a theme that stems from some

01:38:15.800 clinical trial. But if you really wanted to be a skeptic, you would say, nope, that's not the exact

01:38:20.400 same patient. And that's not the exact same situation. And therefore you can talk yourself

01:38:25.320 out of doing anything. And I'm super fond of saying that being a preventive cardiologist is no one should

01:38:33.120 feel sorry for me. I have the best job in the world, but, but it's difficult in that we only know

01:38:36.840 success by the absence of failure. So there's no one who's going to come to me tomorrow and say,

01:38:42.280 gosh, Ethan, thank you for the fact that I've, I'm 46, that I did not have a heart attack again

01:38:47.720 this year. It just doesn't happen. Right. That's a great way to explain it. Whereas the other way

01:38:51.800 around, like I've had a few patients, if you're an orthopedic surgeon, for example, that's right.

01:38:55.700 You break your leg, you fix it or an interventional cardiologist, right? You show up in the cath lab with

01:39:00.340 a STEMI, you know what you did. The outcome is clear. The outcome is not that clear in prevention,

01:39:04.840 unless there's failure. So those examples, and I've had a few recently, I've been public about them

01:39:10.740 on Twitter that are treatment failures, but maybe not personal failures. In fact,

01:39:16.020 I don't think I managed the patients incorrectly, but the fact is they had events while they were

01:39:22.200 under my care. Those live with you for a long time. And so then the question is, I know you're

01:39:25.780 a race car driver. The question is, is your reaction to that to then have a tendency to want to oversteer?

01:39:30.720 So because I have these anecdotes, these very profound anecdotes of young people who've had

01:39:36.060 terrifyingly scary outcomes. And, you know, I was not as aggressive as I could have been,

01:39:42.240 but probably still following the sort of guidelines. Is that going to guide me as a physician to be

01:39:47.140 more aggressive in the future? And again, we're not going to have clinical trial data to help us

01:39:51.800 here. This is all art and judgment. The subtitle of my book, I'm hoping if the publisher lets me is

01:39:57.740 going to be called the science and art of longevity. There's a title to it, but that's

01:40:02.000 the subtitle. And I'm insistent upon that order because normally you say it in the reverse,

01:40:06.180 the art and science of whatever, but it's the science and art. You're informed by science,

01:40:11.080 but in the end, this still comes down to an art. Well, and it's the art of the science too,

01:40:15.120 as you said, right? I mean, it is sort of, how do you put this? And then there's the whole other

01:40:18.320 layer, which is how do you communicate it with your patients? And how do you include them as a

01:40:23.260 partner in making these decisions? I mean, that's where things get really interesting.

01:40:26.560 And there are other physicians, because typically these are not just discussions between two people.

01:40:30.580 So what's a CAC, what's a CTA, and how do you use them?

01:40:33.860 The way I break this down is there's anatomy, so anatomic tests and there are physiologic tests.

01:40:38.560 So physiologic tests would be like a stress test, a treadmill test. So you get up on the treadmill,

01:40:42.840 you walk, and then you have one of a few different ways to kind of determine,

01:40:47.440 in addition to whether you have symptoms, determine whether there's ischemia, meaning there's a

01:40:51.000 difference in the supply and demand of oxygen. So you can use EKG, or you can use an echo,

01:40:55.280 or you can use what's called nuclear, which is basically a radio-labeled potassium analog that

01:41:00.540 lights up living cells. So that's one way, that's the physiology. And that's been the standard of care

01:41:06.660 in this field for 40 years. In the past 20 years or so, people have started to explore whether there's

01:41:14.760 a role for anatomy. So one anatomic test is an angiogram. So you can take somebody to the cath lab,

01:41:19.840 and you can put a catheter up there and inject some dye into the artery, and you can see if it

01:41:23.800 looks like there are blockages. That's not something that you would apply broadly to a population. You

01:41:29.080 wouldn't use that as a screening test. And patients often ask me, in fact, sometimes people will come

01:41:33.480 in the office and they'll say, I should have an angiogram. No symptoms, just I want to have an

01:41:37.400 angiogram. I want to know they don't have any.

01:41:38.780 Right. Just a preventative, get a baseline.

01:41:40.840 Yes. Which is very much like, I want to make sure I don't have colon cancer or breast cancer. And

01:41:46.120 the same things that come up as being difficult and controversial in any screening, whether it's

01:41:52.300 prostate cancer screening, breast cancer screening, any of these screening areas are going to be

01:41:55.360 super controversial here. But I think the community agrees mostly that screening angiography is not

01:42:00.380 appropriate. But the question is, are there other things that could be appropriate in the right

01:42:04.660 people? And in general, the population of people that we're talking about are people in this

01:42:09.680 intermediate risk category. So if you're low risk, for the most part, there's not going to be much

01:42:13.360 benefit to adding one of these tests. We're talking about asymptomatic people here. So 1.00

01:42:17.040 if you're symptomatic, you fall into a different category altogether. So asymptomatic people. And 1.00

01:42:21.200 if you're high risk, you probably fall into a different category as well. And what I tell people

01:42:26.480 is that the results of these anatomic tests that we might do are really about answering one question,

01:42:33.160 which is how aggressive should we be with our medical therapy? And what they are, they are

01:42:39.000 triggers for basically saying yes or no to statins or yes or no to statins and PCS-19 parameters.

01:42:45.900 Basically, it's a way of gauging how aggressive we want to be. So again, low risk, you're probably

01:42:49.240 not going to use a statin anyway, except in sort of unusual circumstances, say somebody has like a

01:42:53.820 very early and extreme family history or something else is funny. But if somebody falls into a low risk

01:42:58.460 category, they probably get left alone. So it's intermediate risk people. So there was a doctor,

01:43:02.800 I believe he was from Houston named Agustin. And he came up with this idea. So people could see on

01:43:09.120 chest x-rays back, you know, a hundred years that if somebody had a blocked artery, that there was a

01:43:15.760 likelihood that you could see the outline of the artery on the chest x-ray. That's because calcium

01:43:19.980 is radio opaque. So people realize that these arteries, again, by autopsies and other conventional

01:43:26.420 imaging, that these arteries get calcified as they develop more plaque. And so this radiologist in

01:43:33.180 Houston figured out that there was a linear relationship between the amount of plaque that

01:43:36.900 you have and the degree of calcium that you have in your arteries. And so he developed a quantitative

01:43:40.960 measure and that's called the coronary artery calcium scan or score. And that is a low dose,

01:43:47.600 low dose radiation screening test that basically just looks for this one thing, which is calcium.

01:43:53.500 And then it quantitatively tells you how much you have. And it then gives you on a percentile basis,

01:43:58.160 what you are, what you should be based on your age and sex. So that's a calcium scan.

01:44:02.840 It's worth pausing for a moment to make sure, because this is a discussion I tend to have with

01:44:06.840 patients a lot, which is what does it mean if it shows nothing, meaning your score is zero? What does

01:44:11.500 it mean if it shows anything that's not zero? And to explain this to patients, I usually have to,

01:44:17.100 again, pull out the pathology textbook and explain all of the things that happen before you would see

01:44:23.120 calcification. So when you look at the stages of lesions, calcification is a very late process.

01:44:28.900 You described it earlier as basically the repair of damage. So if you have a calcium score of zero,

01:44:36.320 does it mean you have perfect coronary arteries? You can't conclude.

01:44:40.980 Yeah, this is a really important point. And to me, there's been a lot of work on sort of what

01:44:46.400 people call the power of zero. So what does a calcium score of zero mean? To me, it's all about the

01:44:50.400 context. So in a 25-year-old or my 15-year-old daughter is going to have a calcium score of

01:44:57.240 zero. It's meaningless for her. It gives you absolutely no additional risks ratification. 0.71

01:45:02.500 Conversely, when a 90-year-old shows up and has a calcium score of zero, it's quite informative.

01:45:09.180 And in fact, I have some patients in their 70s and 80s who I've taken off of statins because they

01:45:15.120 had a calcium score of zero. And in those patients, their risk of having a heart attack is not zero,

01:45:19.720 but it's low 1%. I mean, again, let's remember that the biggest risk factor for having a heart

01:45:25.200 attack is your age. So age supersedes everything else. It supersedes all the other known risk factors

01:45:30.480 that we use every day. Your lipids, your blood pressure, your presence or absence of diabetes.

01:45:34.660 Although I will throw in my rant here. Why is age the greatest risk factor?

01:45:39.480 I don't know. It's a good question.

01:45:41.000 I mean, I would argue, as Alan Snyderman has argued, that it's an area under the curve issue.

01:45:45.740 It's an exposure question. So age is the greatest risk factor for a number of diseases in which it's

01:45:53.000 not entirely clear why. But I think in atherosclerosis, you would have to argue that this

01:45:58.260 area under the curve shows a monotonic progression in an individual, just as we see progression of

01:46:03.940 polyp to cancer in the colon. So that's exactly why I think the older you are, the more interesting

01:46:11.140 it is to have a zero calcium score. It's Bayes theorem at its finest. You are acquiring more

01:46:16.560 and more information as somebody goes on. So, you know, when I see 35 year olds with zero calcium

01:46:21.380 scores saying, see, I'm totally fine. I say, I have no goddamn clue. How do you know that? 0.99

01:46:27.060 But the contrast is that if you see a 35 year old who does have calcium,

01:46:30.540 That's exactly right.

01:46:31.400 That's informative.

01:46:31.720 So there you have, the way I describe it is it's a two by two, and there's only two

01:46:36.420 quadrants that are interesting to me. Young people, meaning someone under say 45 or 50

01:46:42.400 who has calcification and old people, and I'm sorry to use the term old, but older folks 0.99

01:46:48.540 call it someone over 70 who has no calcification. If you're in those categories, Bayes theorem

01:46:54.840 is lighting up like ding, ding, ding, ding, ding, ding, ding, ding. If you're in the other

01:46:58.380 two quadrants, I don't think I've learned anything.

01:47:01.640 Mostly. So I started, again, I'll frame this as I started off thinking there was no value

01:47:05.640 to calcium scans. I've evolved to think, all right, well, there's going to be value in

01:47:09.580 these two quadrants you described. I also think there's probably some value in the extremes

01:47:13.780 in say, I saw a patient recently who was in his fifties, early fifties, who had a calcium

01:47:19.720 score of 1300, you know, asymptomatic primary prevention.

01:47:23.440 I have a 49 year old who just scored 4,500.

01:47:26.560 Right. So that, I mean, that to me is useful, right? So that takes us, again, not going to

01:47:32.000 ever be supported by clinical trials, but that takes us from, okay, well, now we all agree that

01:47:35.520 you should be on at least statin for primary prevention, but maybe we want to be even more

01:47:40.600 aggressive, right? And so that's when you start thinking, this is actually going to change what

01:47:44.360 we do or change, at least in my head, the risk benefit ratio.

01:47:48.820 Anecdotally, I want to tell you about that patient because this is just getting off in the weeds

01:47:52.340 a little bit. This is a patient whose family history, if I shared it with you, you would say,

01:47:58.100 oh, he's going to have an elevated LP little a father had first MI at 42 dead by 49 brother

01:48:04.300 had MI in forties, et cetera, et cetera. When you hear these stories, you say, well, it's elevated

01:48:09.280 LP little a for sure. You do an advanced blood test on them, normal LP little a, and you want to know

01:48:15.520 something else. Lipids aren't even that crazy. It's not like he's got FH. It's not like he's got

01:48:21.200 an APOB in the 99th percentile. Yeah, he's probably in the 70th percentile, but it's like,

01:48:27.980 these are the patients that keep you up at night. And it's like when I was talking to Richard Isaacson

01:48:32.280 about Alzheimer's disease, he's way more afraid of the patients with a ravaging family history of

01:48:38.640 Alzheimer's disease who have normal APOE because there's clearly something going on and I don't even

01:48:44.000 know what it is. And that's sort of how we feel about these patients of which we have many where

01:48:49.040 it's not even like, you know, they're at the 70th or 80th percentile of risk by lipids and by other

01:48:54.800 metrics, but by story and now by calcium and other tests, I mean, they're at the 99th percentile.

01:49:01.520 I had a patient I talked about a lot on Twitter a couple of weeks ago who ended up in his early

01:49:05.320 fifties and ended up having like a very, very tight proximal left main, like just the kind of thing

01:49:09.900 that you're the so-called widow maker. You look at this thing and you think there's no way in the

01:49:13.340 world this guy should have walked into my office and he did. And his first past lipids were all

01:49:19.340 completely normal. You know, his, you know, regular fasting lipid, he had no obvious risk factors.

01:49:23.720 And then we got his LPA back and it was high and I was relieved and I was relieved because on the one

01:49:29.060 hand, I knew I understood what was going on. On the other hand, we're still left with this.

01:49:33.660 Well, we still don't really have a direct therapy to offer, but at least we've got a couple of indirect

01:49:39.300 things till Sam Tamekis' ASOs are around. Did you look at his aortic valve yet?

01:49:43.840 It's normal.

01:49:44.640 It is.

01:49:45.100 Yeah.

01:49:45.520 That's so interesting. I would have guessed someone that young with such aggressive athera would also

01:49:50.140 have something abnormal in his aortic valve. Interesting. All right. So that's the calcium

01:49:55.080 score. I've always found it very difficult to interpret the data on this because they tend to

01:50:00.820 report it without clearly specifying the age cohort. So for example, everybody loves to quote a study

01:50:06.560 that would say a zero calcium score means your 10 year risk is 1.9% or something to that effect.

01:50:12.120 But you can't actually say that without knowing the age of the cohort because a zero calcium score

01:50:16.500 in a group of 30 year olds probably has a 10 year risk of less than 1%. A zero calcium score in a

01:50:22.280 cohort of 70 year olds would clearly be five to 10% still. So that's another thing that I just find

01:50:27.340 very sloppy, but I don't have these discussions too much in social media because I find it so frustrating,

01:50:31.760 but it's that you just don't have the time to explain that nuance to folks.

01:50:34.660 It's a great point. And I use the MESA calculator because it's a nice, you know,

01:50:38.460 it has the traditional risk factors and then you can add in the calcium score if you have it.

01:50:43.220 So MESA, multi-ethnic study of atherosclerosis, one of the two largest cohorts we have ever to

01:50:48.900 study atherosclerosis along with Framingham.

01:50:51.040 And then they'll give you the 10 year risk plus or minus. So they'll say, well, this is what the

01:50:55.000 10 year risk would have been without the calcium. Yeah. And that's somewhat helpful. But again,

01:50:58.700 it comes back to this being way more art than science. I mean, this is at this point in

01:51:02.660 primary prevention. This is still very much art and not science and probably will be for our

01:51:06.400 lifetime. So we'll have to get used to that. So then CT angiograms basically said, look,

01:51:10.180 we're going to do more or less what the angiogram did in terms of providing anatomic detail. Cause

01:51:14.800 the calcium score doesn't really give you anatomic detail of the artery. It's basically just showing

01:51:18.100 you, show me every place where there was a disaster that's been repaired. The CTA says, well,

01:51:23.520 now I'm going to actually show you both the negative and positive image of the artery. So you'll see

01:51:28.100 everything that's going on, but it's in many ways, even a better test than the angiogram because you

01:51:32.220 get to see the wall and you can look for soft plaque or other forms of injury to the artery that

01:51:38.180 haven't yet reached the point of calcification. Assuming somebody doesn't have a lot of calcium,

01:51:42.140 if somebody has a lot of calcium, then you can't really do it. And assuming that you can block the

01:51:45.560 heart down, the scanners have gotten fast enough now that the quality of the data we get back from

01:51:49.420 these is spectacular. And we get this information we never used to have unless you did IVUS. So IVUS

01:51:54.140 is intravascular ultrasound. So when you do an angiogram, you don't get to see what's underneath

01:51:57.880 the endothelium. You have to infer it. So now you can see it. So you can see not just the plaque.

01:52:04.180 So I'm fond. I have a strong family history of melanoma. And I'm told by my dermatology colleagues

01:52:08.960 that nevi melanoma grows down before it grows up. So what's happening underneath the surface,

01:52:14.640 you can't see. And then when it starts to grow, you start to see on the surface.

01:52:18.260 That's right. Luminal narrowing is a very, very late consequence of atherosclerosis.

01:52:22.880 So that's why I said, if we did a CTA on us or an average 50-year-old, you'd see plaque. It may not

01:52:28.220 narrow the lumen more than a couple percent, if at all. It may just look like a normal variation and

01:52:32.980 sort of lumpy bumpiness, but you'll see plaque. And that plaque grows down before it grows up.

01:52:38.540 There's a ton of information. The question, of course, is what is the value of that information

01:52:41.860 and how does it change our management? That's-

01:52:43.960 To me, it just adds color to the discussion we already had.

01:52:46.700 It adds great benefit on the negative and on the positive, depending on the age.

01:52:52.040 But you take that patient with a zero calcium score whose CT angiogram is pristine. Their

01:52:57.180 arteries are enormously patent and there's not a shred of evidence of soft plaque. It's still not a

01:53:02.300 guarantee. I mean, we know that, right? Even IVUS could miss plaque or vulnerability, but it makes

01:53:08.100 you feel a bit more comfortable that your level of aggression can be lower and the converse then is

01:53:12.940 true as well. So then what's the cost, right? So other than the theoretical cost of the radiation,

01:53:18.340 which is probably something and the- Yeah, right now they're going so fast. These things are at

01:53:22.460 a really good scanner. So that the top shelf scanners here are doing CT angiograms at about

01:53:28.420 two millisieverts of radiation. So for the listener, millisieverts, the unit in which we measure

01:53:34.160 radiation, the NRC limits people or suggests an upper limit of 50 millisieverts for a year.

01:53:41.380 We want our patients to be below 20 millisieverts for a year. Living at sea level exposes you to

01:53:48.640 maybe one millisievert over the course of a year. Living in Denver probably exposes you to four or

01:53:54.200 five millisieverts for a year. So that puts the two millisieverts in context. The first generations 0.99

01:54:00.500 of these, Ethan, 20 millisieverts just for a CT angiogram. What's a chest x-ray? Incredibly low.

01:54:06.580 It's less than one millisievert. So we can sort of dismiss that. There's very little risk of

01:54:11.260 the contrast in most people. Obviously if you're older, you have kidney disease, something else

01:54:15.100 then it becomes a different issue. One of the greatest costs is the economic cost. It's rarely

01:54:18.180 covered and it's a, I don't remember exact cost, but it's, it's not trivial. It's probably, you know,

01:54:22.280 $2,000, $2,500. So it's a, it's a quite expensive study relative to the CT, the calcium score,

01:54:27.740 which I got to put a plug in, by the way, your institution here needs to get its head out of its 0.99

01:54:31.580 ass. I mean, Jesus. I send patients to Stanford. Stanford. Stanford's doing CT angiograms 0.99

01:54:37.200 for like a hundred bucks and you coconuts here are charging like thousands of dollars.

01:54:41.540 Don't get me started. Well, I think it'll change now with the new guidelines. I think

01:54:44.440 the price has to come down, but it's such an interesting example.

01:54:46.460 I have to send patients from San Francisco to drive down to Stanford. They're pissed off,

01:54:49.800 but I'm like, you know, I tell my patients, this is the one example of there actually being an

01:54:53.940 efficient market in medicine, right? Because third-party payers don't pay for it up until

01:54:58.560 recently. They haven't paid for it. You have to pay for it out of pocket. So these guys realize the

01:55:02.500 only way they're going to get it, no one's going to pay $3,000 to get it. It's the same basic

01:55:06.420 machine, same everything else. It's just that there's a market.

01:55:09.560 It's a fixed cost is already sunk and there's no variable cost. It's just, yeah, it's comical.

01:55:13.780 The question then is sort of what is the risk of the CTA? And so from my perspective, the risk

01:55:17.980 is with the right patient, the right doctor is zero. The wrong patient, the wrong doctor,

01:55:22.960 it could be high. And so the example I use is I've got patients, and I'm sure you have a lot too,

01:55:26.920 who are anxious and they can't, even though they don't anticipate that they're going to be this way,

01:55:32.040 they can't live with this sort of thing growing inside of them. I'm sure you had

01:55:35.660 patients when you were doing cancer surgery who just basically said, get this tumor the hell out

01:55:41.060 of me. I don't, I don't want cancer anymore. Get it out. And even though, you know, prostate cancer

01:55:45.040 is a great example, right? There are lots of ways to treat prostate cancer today. And some of them

01:55:48.220 involve cutting the tumor out. Others that are equally as effective probably are close don't,

01:55:53.400 but some people mentally can't get their head around living with cancer. And I think I have patients

01:55:57.880 of mine who, when they get a CTA and they see there's plaque there, they just can't sleep

01:56:01.860 because they think this thing is going, I'm going to die from a heart attack. And no matter what we

01:56:06.400 say, no matter what I say, I can't dissuade them of that. That's to me is the one cause. So that's

01:56:12.320 where I would sort of say, this is why I'm not going to, that in the economic thing is sort of

01:56:16.020 why I'm not going to widely do this in everybody, but in certain people, it's spectacular.

01:56:21.740 So before we leave this, let's just talk about what heart flow is, because that's something that's

01:56:25.840 come on the scene in about the past. I want to say it's about three years ago, maybe four years ago.

01:56:30.160 It's a company here in the Bay area, correct? And they're layering on an analysis to the CT

01:56:37.680 angiogram. Can you explain a little bit about what that is and where it came about?

01:56:41.340 I'm going to try and keep it brief because I got quoted in an article over the summer about heart

01:56:45.420 flow. And I think I said something like, I don't wake up in the middle of the night dreaming about

01:56:49.280 which patient I want to use heart flow in. So I don't want to be too harsh on them. This is

01:56:54.460 basically based on the idea that you can use physics to measure the percent narrowing of an

01:57:00.540 artery by measuring the velocity across the lesion. So as you, anyone who puts their finger on the end

01:57:06.120 of a garden hose knows that the velocity of the water that comes out of the garden hose goes up

01:57:09.760 as you narrow the opening of the garden hose. So the same thing is true in an artery. If you have

01:57:14.440 a narrowing, then the velocity across that is going to be faster. And there will be a pressure

01:57:19.360 gradient. So you can actually measure pressure using a catheter with a pressure sensor on it.

01:57:24.680 And you can measure pressure before and after the blockage. And you can basically then infer how

01:57:28.820 severe the blockage is. And that technique was first developed invasively in people who were going

01:57:34.020 into the cath lab. And there were patients where you'd look at the artery and you'd think,

01:57:39.200 yeah, we're looking at this in two dimensions and we can't really see, it doesn't, it's hard to see

01:57:43.760 exactly how severe this is. So there were a couple of these tools that were developed. One of them is

01:57:47.060 this intravascular ultrasound IVUS. And then the other one was this FFR, which is a way of measuring

01:57:52.740 pressure and basically inferring the degree of blockage. So fractional flow reserve is the

01:57:58.000 calculation of P2, which is the pressure beyond the occlusion over P1. Right. And they do it before

01:58:06.120 and after administering a vasodilator. So they'll give a patient a vasodilator and they can measure sort

01:58:11.700 of maximal blood flow. And you can do that in a normal artery. In a normal artery, you'll see it augment

01:58:16.300 in a diseased artery. It doesn't, it goes, it'll go down. And there are two big trials that looked

01:58:22.220 at the use of invasive fractional flow reserve. One was called FAME. The other was called FAME-2.

01:58:27.660 I'll spare everybody, me looking up exactly what FAME stood for again. What did those studies show?

01:58:33.140 The take-home, and again, there's controversy as with almost anything in cardiology,

01:58:37.040 there's controversy. But the take-home was that if you had an FFR that was less than 0.8, so you use a

01:58:41.420 mathematical formula to basically calculate. And if it was less than 0.8, if you then put a stent in

01:58:46.220 that artery, people did better. So it suggested that this could be a useful tool to help stratify

01:58:51.740 who should be getting stents and who shouldn't be getting stents. And I think a lot of our

01:58:54.460 interventional cardiology colleagues have been using it that way for the past few years.

01:58:58.940 Again, there are questions. I don't want to get into the controversy, but there are questions about

01:59:01.940 the technique and about how they did it. And there are some questions about the validity of it,

01:59:06.580 but it's mostly a useful tool and people do use it, but it's invasive. So you have to be having

01:59:11.920 an angiogram to do it. So somebody came up with this idea that you could do the same thing

01:59:15.720 using just a non-invasive CT scan, and you could basically measure the velocity and infer the

01:59:22.360 pressure using Ohm's law. So they did the same thing. They gave a vasodilator knot, and they can do

01:59:29.060 this. Boyle's law, sorry. Boyle's. Yeah. I don't know. You're the engineer. It's been a while.

01:59:34.600 So they did this, and lo and behold, they were able to measure this thing non-invasively. You

01:59:39.860 don't have to have a catheterization. You can do it all through CT. And then HeartFlow basically

01:59:44.900 is selling this software as an add-on package to the CTA, and they'll give you what's called a CTFFR,

01:59:51.060 which is their calculated FFR. Is this changing your practice? Because when it came out, I started

01:59:56.340 doing it, and then I realized, especially just based on Orbita, that it wasn't really going to

02:00:02.620 change my management. So I have not done a HeartFlow study in probably over a year on anybody.

02:00:07.580 That's the point. Whereas the anatomy of sort of whether there's plaque there does help guide you

02:00:12.560 about how aggressive to be with your medical therapy. It's not clear to me that we're learning

02:00:16.360 anything from adding the HeartFlow on. So I have never ordered one. It doesn't mean that I won't

02:00:21.240 order one in the future. I've done a lot of this where I've said I've never ordered one, and then 10

02:00:24.180 years later, I've ordered one. I've ordered a lot. So I reserve the right to be wrong on this one.

02:00:29.100 But my instinct is, like you say, we have the information from these two other big trials

02:00:33.600 that tell us that this is probably not going to be a huge game changer because what information

02:00:39.120 is it adding and how are we going to react to it? So their pitch from a business standpoint is that it

02:00:44.640 saves unnecessary catheterizations. That's a business discussion that I think I generally don't

02:00:49.720 believe right now, but that's their case. And if you're an individual patient, what value is it going

02:00:54.960 to bring to you? Or you as a doctor, what value does it bring to you? I can't think of a big value

02:00:59.500 today unless you had somebody who you were going to send to the cath lab and you thought this would

02:01:04.140 help sort of make you feel comfortable about not doing it. Let's pivot for a second because we're

02:01:09.560 sitting here in your office. We've been nerding out for two hours on preventative cardiology,

02:01:17.200 but I'm looking at your very, very beautiful whiteboard. And not only do I have tremendous whiteboard envy,

02:01:23.620 but there's not a single thing on here that looks like it's about cardiology.

02:01:28.640 If I didn't know better, I would think I was in the lab of David Sabatini or one of my other

02:01:36.280 friends, you know, Luke Hantley. I mean, we've got insulin receptors, PI3K, we've got mTOR. You've

02:01:43.900 even delineated between mTOR complex one, two pathways to autophagy, lots of growth hormone, IGF

02:01:51.200 pathways, pancreas, liver. Are we in your office? Whose office is this?

02:01:57.900 I came here to San Francisco 21 years ago to study in the lab. The person who became my mentor was my

02:02:03.420 boss until he left to go work in Novartis a couple of years ago, a year ago, two years ago,

02:02:07.360 Sean Coughlin. And I came here because I was interested in the biology of blood clotting.

02:02:11.000 And so when I left his lab and decided that I want to stay here in San Francisco, I had to find 0.92

02:02:14.600 something that I wanted to do that would be not what he was doing because that would be dumb.

02:02:18.380 And so at that time, I got very interested in the biology of sex differences. And I don't want to 0.85

02:02:24.880 spend seven hours describing it, but it's a really interesting literature. And so I did a deep dive

02:02:30.160 on it and thought, gosh, there's a lot of reasons to think that there may be significant differences

02:02:34.100 in the way men and women behave clinically and biologically with regard to the clotting system.

02:02:40.400 And there may be reasons to expect why that would be, right? We talked a little bit before about the

02:02:43.960 placenta and pregnancy. Men don't have to have this other thing living inside of them and therefore

02:02:51.180 could afford maybe to optimize more towards a sort of wound healing defense strategy, right?

02:02:56.940 Men, there's a difference in sort of the way you might, if you were to design a system,

02:03:01.180 if you were to create evolution there, you may be able to optimize it, this system for the two sexes

02:03:05.880 differently.

02:03:06.240 That is the most interesting idea I've ever heard because one of the questions I've always

02:03:11.100 batted about in my mind is all things equal. Why are women less susceptible to cardiovascular 1.00

02:03:16.060 disease? Or at least I wouldn't say it that way. Why do women experience a phase shift? They get it 1.00

02:03:20.540 later. And you look at blood pressure differences, maybe women have lower blood pressure. Another one 1.00

02:03:25.860 is iron. Women have less iron for 30, 40 years of their life. They're basically doing a blood donation 1.00

02:03:32.080 every month. I never once considered what you just said, which as you say, it actually makes a lot of

02:03:38.400 sense. Men would in theory evolve to heal from a wound much quicker, even at greater cost down the line

02:03:47.020 because of their role in a hunter gatherer society. Whereas the woman must evolve to protect the 1.00

02:03:53.660 offspring much more. And you could see how that would be a lower immune response, a lower inflammatory

02:03:59.240 response, a lower pro thrombotic response. That's, that's an, I can't believe I didn't even know that.

02:04:04.920 Well, if you want to get really crazy, you can start to imagine, well, maybe women evolved to be 1.00

02:04:08.800 the less sort of masculine of the sexes. In other words, maybe they're the ones not going out and

02:04:14.660 throwing spears and fighting because they have to carry pregnancy. And so they, they're a little bit 1.00

02:04:20.820 more susceptible to bleeding, you know, and you know, there's the added sort of complication of

02:04:25.900 menstruation, all these other things. So anyway, I thought, wow, this is a really interesting

02:04:28.740 area. And the, the reason it was particularly interesting was that in the work that we had

02:04:33.000 done primarily in mouse genetic models, it looked like there was a pretty significant difference

02:04:37.060 in the rate of clotting between male and female mice. So I thought, well, this would be a really

02:04:42.220 fun thing to study. And of course I did exactly what you did, which was, I immediately kind of

02:04:46.440 latched onto what must be sex hormones and did some early kind of pilot studies looking at the role

02:04:51.900 of estrogen receptor different. And it didn't pan out. And so I did more reading about sex dimorphism

02:04:59.460 in general. And it turns out that mammals have this remarkable sexually dimorphic liver. So if you

02:05:05.520 take the liver out of a, of a mammal, human or mouse, there's a set of genes that are expressed

02:05:10.740 dramatically different, even up to a hundred fold different males and females. And a lot of these genes

02:05:16.180 are genes that regulate sex hormone, sex hormone binding or modification. Others are sexually

02:05:21.460 dimorphic for reasons we don't understand. But if you do a sort of array of gene expression,

02:05:25.840 whether you're using an old fashioned chip array or whether you're doing RNA-seq, you see this

02:05:31.420 tremendous signature difference between the genes that are turned on in the liver of females and males.

02:05:36.600 Well, where are clotting factors made the liver? So I thought, well, this is what we're going to do.

02:05:42.120 And so I set off to like basically try and understand if how this happens. So what regulates

02:05:48.700 the dimorphism of liver gene expression in mammals? It turns out it's not sex hormones. That's actually

02:05:55.020 secondary. It turns out that it's a dimorphic pattern of growth hormone secretion. So it was

02:06:00.300 described back in the fifties and sixties that there was a factor from the pituitary called

02:06:04.580 feminizing factor, but it wasn't known what it was. If you just ground up the extract of the pituitary

02:06:09.640 and injected it into a male or female mouse, you could basically drive the expression of these

02:06:14.580 genes. And ultimately it was found in a beautiful paper. I think it was published in 1982 in Cell by

02:06:20.660 Richard Palmer's group. It was found that that factor is growth hormone. And that if you...

02:06:25.100 Which by the way is the last one I would have predicted. Like I would have said,

02:06:28.120 oh, it's probably luteinizing hormone or follicle stimulating hormone. I wouldn't have guessed

02:06:31.920 growth hormone.

02:06:32.260 No, it's growth hormone. So males have a more pulsatile pattern of growth hormone secretion. So they have

02:06:36.880 longer intervals between the pulses. Females are more continuous. And so they have very short... 0.96

02:06:41.780 So the guys have these spikes that are less frequent and the women are putting... Now area 1.00

02:06:46.840 under the curve, about the same.

02:06:48.320 Mean is the same. So what's cool is you can actually take an intact, normal wild type male

02:06:52.380 rodent and put a pump in and deliver continuous growth hormone. And you can completely reverse

02:06:58.120 the signature of their gene expression in the liver. And you can do the opposite with a female 1.00

02:07:01.960 mouse. You take a female mouse and it... Because if you give exogenous growth hormone, 1.00

02:07:05.500 you basically suppress endogenous growth hormone secretion from the pituitary, you can give a

02:07:09.760 couple of injections. I think it's just one or two a day to a female mouse and you masculinize 1.00

02:07:14.120 the liver.

02:07:14.960 We live in a world where a lot of people have decided it's a good idea to prescribe growth

02:07:18.600 hormone to reduce the effects of aging. I talked about this very briefly with Nir Barzilai on our

02:07:24.200 podcast. And I would say that Nir's view from that was that there might be some benefit in men in

02:07:30.220 certain circumstances and women less so. But is it safe to say that if a man and a woman

02:07:35.020 are both receiving the same dose and same dosing pattern of exogenous growth hormone,

02:07:40.460 you would evaporate that difference of gene expression in the liver? Because you would

02:07:45.460 have suppressed the pituitary secretion?

02:07:47.340 Yes. You should.

02:07:48.700 Have you done that experiment?

02:07:49.900 In people?

02:07:50.560 No. In mice?

02:07:51.360 Yeah.

02:07:51.700 Okay.

02:07:51.940 Oh, yeah, yeah. Yeah. We did that experiment. So I got into this because I was interested in

02:07:55.180 clotting. And that makes sense as a cardiologist. We just spent two hours talking about why clotting

02:07:59.740 is interesting to a cardiologist. So starting off my lab as a junior assistant professor,

02:08:03.640 I thought, well, this is what I'm going to do. I'll understand the biological basis of sex

02:08:06.840 differences of clotting, get into this growth hormone pattern. And so we started to do that.

02:08:10.760 And we, as a lot of people of my generation do, science is a little bit of a game of you do

02:08:17.120 what everyone else does and see what comes out. And so we started knocking out different

02:08:20.000 components of the growth hormone signaling pathway in mouse. And we got to this protein called

02:08:25.020 JAK2, which is a kinase, which basically adds a phosphate group to proteins. And it's important

02:08:30.940 for other areas of medicine and science, but it's an obligate signal transducer of growth

02:08:34.800 hormones. So downstream of the receptor, growth hormone binds JAK2 basically allows this signal

02:08:41.220 to be transduced.

02:08:42.060 Is there a different receptor for growth hormone in the liver versus outside of the liver?

02:08:45.600 No. GHR is the same everywhere. There are different isoforms. I think there's a soluble

02:08:50.940 one, which is basically a GH binding protein, but the GHR is the same. And that signal is transduced

02:08:57.300 because growth hormone receptor itself is a type one cytokine receptor. It doesn't have

02:09:00.640 any intrinsic tyrosine kinase activity. It has to partner with this other non-receptor

02:09:06.060 tyrosine kinase, JAK2, to transduce the signal. So we eventually knocked out JAK2 in the liver

02:09:11.660 and hepatocytes. And I'll never forget that I had a technician working in the lab at the

02:09:16.220 time. Actually, he was the only person working in the lab at the time. It was just the two of

02:09:19.460 us. And he came to me, he said, Ethan, there's something wrong with these livers. I don't even

02:09:23.640 need to genotype them. And I'll show you a picture, but it turns out that the livers in

02:09:28.060 the animals where we had selectively deleted JAK2 in hepatocytes were basically turned into

02:09:34.280 atomized tissue. It was the most- So you were creating NAFLD?

02:09:37.360 Yes.

02:09:38.200 So just let me make sure I understand what you're saying. If you knock out JAK2 in a liver,

02:09:43.260 you prevent growth hormone from exerting its transduction in the liver because you need JAK2

02:09:49.460 as that second tyrosine kinase. So it's basically like taking away growth hormone from the liver.

02:09:55.880 That's right.

02:09:56.360 And if you do that, you accumulate fat in the liver.

02:10:00.020 25-fold increase in-

02:10:01.400 25-fold. So it becomes clinically grotesque. I mean, you just basically create foie gras.

02:10:06.260 In fact, there was a moment in time where I thought, well, this would be a fun

02:10:08.680 side business. I'll just make foie gras because we can engineer it.

02:10:11.880 Quick question. Did the triglycerides go up or stay the same?

02:10:15.460 Serum?

02:10:16.080 Yes.

02:10:16.560 No, they were the same.

02:10:17.700 So in other words, you shut off export, it seems like, significantly, right? There was no flux.

02:10:22.860 There's no outward. You were basically not getting any VLDL or triglyceride out of those livers.

02:10:26.740 Well, so we didn't know what we did at the beginning. We had no idea. In fact, I took the

02:10:30.540 liver to a pathologist because I said, I don't know what this is because I didn't know what it was.

02:10:34.480 And he said, Ethan, I'll look at another microscope for you, but I know what this is. And I said,

02:10:37.720 what's that? He said, it's NAFLD. And I said, what's NAFLD? And he said, it's non-alcoholic fatty liver.

02:10:41.360 And I said, okay. And so then of course, I started with the very basic,

02:10:45.420 like, all right, well, what are the causes of NAFLD? It's an increase in synthesis,

02:10:48.940 an increase in DNL, it's a decrease in export, an increase in uptake, or it's a decrease in

02:10:52.700 oxidation. So we started to go through this whole list. And to make a very long story short,

02:10:57.240 what we settled on and published, we published a model that I think mostly we've validated to

02:11:02.880 be true, although there's a little bit of controversy about it. And it comes back to what

02:11:06.460 you said. So growth hormone signals, and one of the products of growth hormone signaling in the

02:11:10.960 hepatocyte and other cells is this protein called IGF-1, insulin-like growth factor one.

02:11:16.260 And 95% of circulating, if not more, of circulating IGF-1 in the plasma. So if you take blood of you or

02:11:22.660 me, it comes from the liver. So it's liver derived. So if you knock out growth hormone signaling in

02:11:26.820 hepatocytes, you completely shut off and shut down circulating IGF-1. There's still local IGF-1 in

02:11:33.800 cells and tissues, but circulating IGF-1 goes to zero. And circulating IGF-1 is the plasma biosensor for

02:11:39.800 growth hormone. So growth hormone's half-life is extremely short. So we evolved a system to measure

02:11:44.660 growth hormone levels, not by measuring growth hormone itself, but by measuring IGF-1. So when

02:11:48.500 IGF-1 levels fall, the hypothalamus pituitary sense that, and then they turn on growth hormone.

02:11:53.880 That is so brilliant. Just as an engineer, I have to sort of reflect on that for a moment.

02:11:58.480 Every time I hear something like what you just said, I'm so glad that I was not in charge of

02:12:03.660 evolution. Because I would have screwed that one up. I would have said, make the growth hormone 0.98

02:12:08.920 sensor sense growth hormone. And of course you would get into a cyclic amplified response because

02:12:16.680 of how short it sticks around. But instead, luckily evolution was in charge and not me.

02:12:21.940 It said, no, no, no, no, no, dude, that's going to be way too fluctuating. Let's look at something

02:12:26.620 that's much more stable. That's a readout of growth hormone. Oh, Hey, how about IGF and have that be the

02:12:31.780 feedback loop? Goodness gracious. I love when nature is so smart.

02:12:36.640 The fascinating thing is that circulating IGF-1, I won't say it's all it does because it does a few

02:12:42.600 other things, but the vast majority of its role is as a plasma biosensor for growth hormone. So it

02:12:47.200 regulates growth hormone secretion. So what happens if you knock down IGF-1, you get this huge increase

02:12:51.480 in growth hormone secretion. You basically get acromegaly, but you get selective growth hormone

02:12:54.860 resistance in the hepatocyte. So you probably remember the growth hormone has been described for a long

02:13:00.480 time forever and ever to be pro-lipolytic. So it's actually one of two hormones that's both

02:13:05.040 catabolic and anabolic, but it turns on, it's a way to mobilize fat from adipose tissue stores.

02:13:10.420 Testosterone being the other, where you can be anabolic to muscle, catabolic to fat.

02:13:14.380 And thyroid hormone, I think is another one where you can be both.

02:13:16.800 Unlike insulin, for example, which would be anabolic to both. Yep. Or cortisol, which would be,

02:13:21.280 actually cortisol is the reverse. It's catabolic to muscle, anabolic to fat.

02:13:25.600 That's right. Yeah. So in any case, if you give an animal growth hormone,

02:13:28.800 if you have a patient who has a tumor that secretes growth hormone, what we disease called

02:13:32.480 acromegaly or gigantism, right? Andre the giant had gigantism. Those people have a decrease in lean

02:13:38.680 body mass, increase in muscle mass. That's basically because growth hormone mobilizes fat from adipose

02:13:43.600 tissue stores. So we reasoned that what was happening was that growth hormone was turning on lipolysis and

02:13:49.320 mobilizing all this fat and that that fat was getting taken up by the liver. What was interesting was we

02:13:53.400 did a relatively deep analysis of sort of all the mouse models that have been made involving the

02:13:58.320 receptor, GHR, downstream signaling components, including JAK2 and another one called STAT5.

02:14:03.860 And we looked at animals that have been engineered to be missing any of those components in the whole

02:14:07.700 animal versus just in the liver. And what was interesting was that the animals where you knock

02:14:11.960 out the pathway in the whole animal had a one to two fold increase in lipid in their liver,

02:14:18.000 very modest, if hardly noticeable. Whereas if you knock out the receptor or this other molecule,

02:14:24.040 STAT5 or JAK2, selectively just in hepatocytes, you get this 20 to 25 fold increase. And so that

02:14:29.680 told us it's not just about having high growth hormone levels. It's about high growth hormone

02:14:34.360 levels and those growth hormone levels acting on the periphery. It's also, there must be some other

02:14:39.740 factor that is important here. And so we reasoned, we did what people do, which we did a gene expression

02:14:45.460 array. We saw there's a molecule called CD36, which is also known as fatty acid translocase.

02:14:50.860 And we saw that it was increased by 20 fold in these animals where we knocked out JAK2 in the

02:14:55.800 liver. And we thought, well, gosh, this could be facilitating uptake. And so we did a series of

02:15:01.620 following experiments after that, where we knocked out CD36 just in hepatocytes and showed that we

02:15:05.340 could make that get better. So we put together this model where growth hormone, basically growth

02:15:10.220 hormone secretion is disinhibited at the level of the hypothalamus pituitary. You get an increase,

02:15:14.640 you get an increase in activity in peripheral tissues. It's then mobilizing all this fat. And then

02:15:20.620 the fat is being taken up sort of preferentially or augmented rate of uptake by the upregulation of

02:15:26.820 this molecule. Now, if you block JAK2 and block the expression of CD36, do you eliminate the fat

02:15:34.760 accumulation in the liver? Yeah. Yeah. That's the experiment we did. It wasn't 100%,

02:15:37.780 but we basically brought it back down to almost, you know, to instead of it being 100 grams per

02:15:43.160 milligram of tissue or something, it was like 20 where a normal mouse would be 11 or 12. So it's

02:15:48.260 near normalized. We also then did an experiment where we said, well, what happens if we block

02:15:52.880 this pathway, the growth hormone signaling pathway in adipocytes, since we're presuming that this must

02:15:57.580 be a product of lipolysis. So what happens if we block this pathway in adipocytes? So we went and

02:16:02.580 made an animal where we knocked out JAK2 in adipocytes selectively. And then we crossed the two

02:16:07.700 together and asked what would happen. Basically, can you reduce the amount of lipid in the liver by

02:16:12.120 knocking out both at the same time? And we did. And that experiment we did almost 10 years ago. And

02:16:18.740 what was so fascinating about that experiment was we never intended to get into any of the insulin

02:16:23.040 glucose homeostasis stuff at all. But when you're doing these experiments, they're expensive. And so

02:16:27.920 we did a bunch of things, you know, measured. We did, I think we did an insulin tolerance test for

02:16:31.720 some reason on these mice. And we saw that the mice where we knocked out JAK2 just in the adipocytes

02:16:36.240 had the most profound insulin sensitivity. Like we actually killed our mice because they died

02:16:41.020 from hypoglycemia. We gave them the same amount of insulin that you gave a normal mouse and they

02:16:45.280 bottomed out and we couldn't rescue them. And they, they were super insulin sensitive.

02:16:49.340 And this was mostly glucose being disposed into the muscle?

02:16:52.760 No, it turns out. So we thought that was interesting. So we then went on to kind of dig

02:16:56.380 deeper on what was going on. Again, here we'd knocked out this pathway just in adipocytes.

02:17:02.000 And we saw that if you knock out the pathway in adipocytes.

02:17:05.060 But I'm confused. So you knock out the pathway in adipocytes,

02:17:08.220 when they're hypoglycemic, you're putting that glucose very easily into muscle or into liver. I

02:17:14.960 mean, you'd have to put it into one of those two, wouldn't you to then, even if you wanted to then

02:17:19.080 put it into fat, into the fat cell, right? Yeah. So, and what you're asking is a phenomenal

02:17:23.780 question. And we had initially, I think naively thought that this was an adipocyte autonomous

02:17:28.260 effect that we thought must be the adipocyte, but then there's not that much glucose that's

02:17:33.220 actually disposed into the adipocyte on a per mass basis. So we thought that doesn't make sense.

02:17:36.620 So we ended up doing what you do, which is to clamp the animals. And it turns out that the

02:17:40.880 defect is almost entirely in turning off EGP. So basically what we do is we completely turn off

02:17:47.920 endogenous glucose production from the liver. Which is really not an insulin sensitivity

02:17:51.620 issue at all. It just masquerades as profound insulin sensitivity, but you basically clamp the

02:17:57.500 portal vein metaphorically and you kill an organism in five minutes if you stop hepatic glucose output.

02:18:03.600 How quickly did you see these animals die? It was within 30 minutes.

02:18:06.980 Oh my God. Yeah. Yeah. It was really profound. And, and so this then sent us off in this like long,

02:18:13.060 what's now been a 10 year journey to try and understand how growth hormone regulates

02:18:16.520 insulin sensitivity or insulin glucose homeostasis and how that happens through the adipocyte. We won't

02:18:22.680 get into all of it now, but it turned me onto the kind of remarkable figure. One of these figures in

02:18:28.040 history that I didn't know before I got into this, that I am surprised he, this guy, Bernardo Husay,

02:18:34.420 who was an Argentinian physician scientist who won the Nobel prize.

02:18:37.840 First Latin American to win the Nobel prize, right?

02:18:40.140 I mean, this guy was unbelievable. I'll send you a link. There was a period of time in the 1930s. I

02:18:45.040 think it was 1936 where he was the first or only author on 14 articles in the New England Journal of

02:18:54.600 Medicine over five weeks. At one point he had, it was the most unbelievably productive period of time

02:19:00.000 I've ever seen. And he ended up winning the Nobel prize. He shared the Nobel prize with the Corys

02:19:03.980 of the Cori cycle, which I didn't know until recently, I was reviewing some of this stuff for

02:19:08.680 our talk today. And I didn't realize that they had collaborated on some stuff back in the 1930s,

02:19:13.680 but he defined the diabetogenic nature of growth hormone. So he did these experiments that I think

02:19:19.420 were really cool and basically got left behind because no one really paid attention to any of this stuff

02:19:23.560 for a long time. But he defined that if you take an animal, the first set of experiments he did was

02:19:28.760 he took an animal and took the pituitary out and then gave the animal insulin and saw that the

02:19:35.680 animal was, he described it as they were more susceptible to being to the toxic effects of

02:19:39.960 insulin. So they bottomed their blood sugar out. He then did the opposite experiment where he took

02:19:44.300 an animal, he lopped out the pancreas and then saw that if he got rid of the pituitary gland,

02:19:49.880 that they were less susceptible to the hyperglycemia, that they basically had improved

02:19:53.360 diabetes. And those experiments continued on for 60 years.

02:19:57.600 Let me just make sure I understood. Take away growth hormone and you're going to become much

02:20:04.540 more theoretically or appear insulin sensitive, but in reality, much more hypoglycemic.

02:20:09.340 Take away growth hormone in the absence of the pancreas and you actually tolerate hyperglycemia

02:20:15.920 better.

02:20:16.480 That's right. And in fact, there were people in the sixties, I believe, who were doing

02:20:20.640 hypophysectomies in people with very brittle type two diabetes.

02:20:24.640 So severing the link between the hypothalamus and the pituitary.

02:20:28.040 They were taking the pituitary right out. Wow.

02:20:30.560 And it was a very effective treatment actually, except that it had significant morbidity and

02:20:35.280 mortality of the surgery. So it was at that time they were, they had perfected the art of

02:20:39.000 taking the pituitary out.

02:20:39.980 But even when they did, you're saying even now that they'd have to replace all of the

02:20:43.240 thyroid and sex hormone and ACTH, the cortisol related stuff, it still turned out to be better

02:20:49.460 off all things considered in how well it managed their diabetes.

02:20:52.360 The ultimate killer experiment happened in 1983 ish or so. And so they took patients with type

02:20:59.360 one diabetes who had an insulin pump. So they were getting a fixed dose of insulin and then

02:21:04.020 they gave them, I believe one or two injections of growth hormone. And you could see the insulin

02:21:09.480 level go like this straight up. Basically their insulin use within a day spikes and stays up until

02:21:16.960 they stopped the GH injections and it comes back down. And in fact, three out of the seven people

02:21:21.140 in this one study got profoundly hypoglycemic when they stopped the growth hormone. So it turns out

02:21:26.400 that there's a drug that blocks the growth hormone receptor that's used as a second line treatment for

02:21:30.300 acromegaly. It's called somovir or picvisimum. It's a analog of native wild type GH just has a few

02:21:38.080 mutations in it that renders it basically as a dominant negative. So it doesn't allow, basically shuts off

02:21:42.800 the receptor and it's used again clinically. And if you look at the package insert, the black box

02:21:48.840 warning is if you take any diabetes drugs, let your doctor know because your risk of hypoglycemia is so

02:21:54.740 high. So it actually reminds me a lot of the sort of low carb ketogenic diet in the sense that it has

02:22:00.140 that same effect. And in that 1983 New England journal paper, they actually talked about growth hormone

02:22:05.520 secretion being as much the cause of brittle type 2 diabetes as the effect. And for reasons that I

02:22:13.060 don't understand, people just forgot about it. And so we set out to kind of try and understand at least

02:22:18.340 at the beginning at a cellular level, how is this regulated? And we're trying to dig down. It turns out

02:22:23.500 that it all comes back to your favorite molecule, mTOR. And it looks like it converges in this intersection

02:22:29.200 between these two signaling pathways that evolved actually only, I mean, growth hormone doesn't

02:22:34.200 exist pre-telios. So there's no growth hormone in flies or worms. It's only in invertebrates and

02:22:40.240 teliosts. And it co-evolved basically from insulin, which is not that surprising, right? IGF-1

02:22:45.900 is basically the target molecule of GH. It's not surprising that GH and insulin would have evolved

02:22:52.580 in a similar way, even though they look very different, they act very different. So the way I think

02:22:57.360 about it is that growth hormones kind of retained a lot of the sort of metabolic effects of insulin

02:23:02.320 and insulin obviously has retained a lot of the growth promoting effects that used to be with growth

02:23:07.880 hormone when they were all one molecule. And so the point is that I think that these two molecules act

02:23:12.620 in concert. And from our experiments, it looks like that's happening in the adipocyte and that basically

02:23:17.900 growth hormone is acting as a break on insulin signaling and that that's controlling metabolism in the

02:23:23.900 whole body at a distance. And so that's what we've been working on. And again, we could probably

02:23:27.180 talk about that for three hours, but it's been keeping us busy.

02:23:31.880 So let me kind of think through this a little bit because it's so interesting. Let's go back for a

02:23:35.960 second to, you didn't mention IGF binding proteins, but where do they fit into this? The IGF-BPs as

02:23:43.400 they're called are also made by the liver, correct? So IGF is mostly made by the liver. IGF-BP-2,

02:23:49.600 BP-3 also made by the liver. These are interesting to most people listening to this because if they

02:23:55.820 pay attention to any of the sort of epidemiology around longevity, there's so much controversy

02:24:01.040 around this idea of IGF and its association with mortality and you have such conflicting

02:24:07.460 information, that's what makes this so challenging, right? So you have these people that have mutations

02:24:12.160 in either the GHR or low GH secretion and or low IGF due to a deficient GHR on the liver and they seem

02:24:22.700 to live longer or at least get less cancer, but then they also seem to get a bunch of other diseases

02:24:27.120 that are kind of weird. But when you look at all cause mortality, it seems to nadir at the 60th or

02:24:34.680 70th percentile for everything. But when you do it by disease, certain diseases seem to get better with

02:24:41.080 higher IGF, like Alzheimer's disease. And I believe cardiovascular disease, you would know this,

02:24:46.180 but others get worse, like cancer starts to go up.

02:24:48.640 So how does all of that make sense in terms of what you're learning at this level?

02:24:55.240 Well, I'll say that the GH meetings are fascinating because they're basically,

02:24:59.400 the room aligns in these two camps. There's one side of the room that believes that GH is the elixir

02:25:04.460 that will promote longevity. And there's the other side of the room that believes that too much GH

02:25:09.600 signaling will negatively impact longevity. It's really interesting the way these two things. I think

02:25:14.200 for me, if you start with what we know from human genetics and animal genetics, it's incontrovertible

02:25:21.420 that excess GH signaling, at least at the levels that you see with transgenic animals or acromegaly,

02:25:28.020 is one of the most potent ways to die, right? I mean, people with acromegaly, untreated acromegaly,

02:25:32.820 die in their 30s and 20s and they die of cardiovascular disease. So I think that's

02:25:36.720 gain of function. And again, the animal's the same thing. It's a great way to shorten lifespan in any

02:25:42.180 animal, whether it's a worm or a fly or increased signaling through this pathway and you'll shorten

02:25:47.000 lifespan. And then the flip side, so you mentioned Lerone patients, right? Those patients don't have

02:25:51.680 increased longevity per se, but they sure do have a decrease in their risk of metabolic disease. It's

02:25:58.680 a much decreased risk of developing type 2 diabetes despite an increase in body fat, right? So they have

02:26:04.400 this body composition where they have a 50% body fat, which you'd think would be dangerous. It turns out

02:26:12.020 that they and the mice, so there's a mouse model of Lerone syndrome. And that's actually the longest

02:26:15.920 lived mouse that's ever been engineered. There was a prize that, I don't know if it still exists,

02:26:20.060 but it's called the Methuselah Prize. And they used to give out, I think it was $5 million to an

02:26:24.620 investigator if you could engineer the longest living mouse that's ever been made. And so this guy,

02:26:29.100 Andre Bartke at Southern Illinois, engineered the mouse that's missing. Actually, he and John Kopchick

02:26:35.060 did it together, this experiment where they engineered the mouse that's missing GHR. And that mouse lived to be

02:26:39.220 1,819 days, which is basically the equivalent of like 207 human years. It's still to this day, the longest

02:26:45.620 living mouse that's ever been made. And that meant it was without GHR in the liver? No, anywhere.

02:26:52.080 Anywhere. Anywhere. And that means for its entire life? The whole life. So these are short and fat mice

02:26:56.560 and they live forever. So, but they have all the other problems that you'd have if you're missing

02:27:01.140 growth hormone congenitally. So they're short and they're, they're pretty fat. What's interesting is that

02:27:06.140 that fat. Are they cognitively the same? The mice or the people? The mice. Well, I guess both,

02:27:11.940 but I was asking specifically about the mice. I think they are. It's a great question. I don't

02:27:14.380 know how much has been done on their cognitive abilities late in life, but, but they clearly are

02:27:18.780 very resistant to these diseases that kill mice in labs. They're healthy by every stretch. They're

02:27:24.720 extraordinarily insulin sensitive. I mean, again, same thing that was described before we ever did these

02:27:29.180 experiments. What we demonstrated was that the insulin sensitivity is conferred by the absence of the

02:27:33.600 signaling pathway in the adipose tissue, which is acting at a distance mostly on the liver.

02:27:38.660 So what do you think confers the protection to the Leron's patients?

02:27:42.040 I don't know. It's one of the things I think about every day, but at a very simple level,

02:27:44.880 I think of it as an improvement in insulin sensitivity, decrease in insulin and decrease

02:27:49.580 in IGF-1. I can't get better than that right now, but I think the circulating insulin levels

02:27:55.540 are nearly undetectable and mice in labs die of cancer. And I think there's probably just less

02:28:02.740 insulin, less IGF-1, and they just don't get cancer.

02:28:06.120 Because this relationship between hyperinsulinemia and high glucose, you and I both are on an

02:28:12.500 advisory board of a company called Virta Health, and that's how we actually met a few years ago.

02:28:16.920 And I remember in one of the meetings, we were discussing this idea that people need to

02:28:22.660 understand the difference between managing glucose and managing insulin, right? If you manage glucose

02:28:28.240 levels, if you keep glucose levels in the ideal range, by any means, you control microvascular

02:28:34.400 disease. You're less likely to go blind and less likely to have kidney failure and have an amputation.

02:28:40.160 But if you do that with a strategy of high insulin, you trade it for macrovascular disease,

02:28:46.040 atherosclerosis, coronary artery disease throughout the body, obviously, you know, in the coronary

02:28:51.080 arteries and cerebral arteries, even the aorta, the large vessels, suggesting that even high insulin by

02:28:57.360 itself can be problematic. So these patients are walking around with very low insulin, which you

02:29:03.220 could see would protect them at the macrovascular level. Because of this, I still think of it as a

02:29:09.220 paradox because I'm still struggling to wrap my head around their insulin sensitivity, how they can be

02:29:14.060 so insulin sensitive, but they also obviously have normal glucose levels. So their microvascular stuff

02:29:19.580 is okay. And then on top of that, you're saying if insulin and IGF are pro, from a metabolic

02:29:26.800 standpoint, pro-oncologic or pro-tumor, you also get that benefit. The one I've never understood is,

02:29:34.520 aren't these things important neurotrophically? I mean, isn't IGF and GH important in the brain and

02:29:39.460 how are they avoiding those consequences? One of the foibles of doing what we do is that we look at

02:29:45.480 what we look at and don't look at anything else. So again, I set out to do these experiments

02:29:49.560 originally because I was interested in the clotting. It was only an accident that we noticed that the

02:29:53.140 liver looked like foie gras. We've never looked at the brain. And I don't know how much others have

02:29:58.220 looked at the brain. And I'm not sure how much what happens in a mouse brain would guide us,

02:30:02.700 but it's a great question. I think if you're just asking about the things that I can measure,

02:30:08.520 so metabolic disease, cardiovascular disease, cancer, and lifespan, it seems like everything

02:30:14.220 aligns towards too much GH being bad and less GH being better. But I can't tell you about the

02:30:21.180 neuro stuff. It could be the opposite there. And why is it that if the patients with Laurent

02:30:25.320 syndrome are less likely to die or they get to phase shift and delay the onset of cardiovascular

02:30:30.900 disease, neurodegenerative disease, and cancer, they don't have a survival advantage?

02:30:35.400 Well, the story that I, I mean, again, you're talking about pockets of a few hundred people. I think

02:30:39.420 the one that's gotten the most attention is this one in Ecuador and there are 300 patients there.

02:30:43.000 I think what's amazing is that the 30-year follow-up of that cohort of people, there was

02:30:47.100 not one single case of cancer or diabetes over 30 years in 300 people compared to like, you know,

02:30:52.880 15% in their age-matched relatives. What I heard was that they were depressed and had an increased

02:30:59.100 risk of alcoholism because they were basically ostracized for being fat dwarfs. I mean, I hate to

02:31:05.100 be harsh, but I think that was my interpretation.

02:31:06.980 Yeah. And I also heard there was a higher incidence of accidental death and other things, which you could

02:31:10.920 argue, well, you have to die of something. So if accidental death explains it, that should still

02:31:16.140 show a survival advantage.

02:31:17.580 I think we're getting into the sort of trouble of trying to-

02:31:20.160 Yeah. Epidemiology just makes this very difficult.

02:31:22.300 Especially in these small like inbred communities. I think it just becomes almost impossible. So 1.00

02:31:26.100 we talked about experiments we're never going to have answers to. We're never going to have a human

02:31:29.500 longevity lifespan study that you or I will see the results of. And we just have to understand that.

02:31:34.960 So, but in the meantime, we can talk about the other stuff. And I mean, the insulin thing to me is

02:31:39.500 interesting because I think there are some people out there in the Twitter sphere who are sort of

02:31:44.160 fall into the category of statin denialists. And I think there's this narrative that's developed

02:31:49.180 out there that statins don't treat the underlying cause of cardiovascular disease, that the underlying

02:31:52.900 cause of cardiovascular disease is insulin. And while I'm firmly in the camp and will ever expect

02:31:58.780 to remain in the camp, that lipids are fundamental and probably the most important driver of cardiovascular

02:32:02.800 disease, of atherosclerotic disease and events, I do think there's good evidence that insulin

02:32:07.780 actually does play a role. And in fact, there were some studies that were published, you know,

02:32:11.640 over the years that showed that, that fasting insulin was an independent risk factor for

02:32:15.980 cardiovascular disease. And it makes sense.

02:32:18.240 Don't we also see this borne out in the type one literature that stratifies patients by insulin use?

02:32:22.740 Yeah.

02:32:23.320 I think James O'Keefe published a paper on this about two years ago. Do you know James?

02:32:26.780 No.

02:32:27.520 Proventional cardiologist in Kansas. Mostly his work is in exercise and the role of too much exercise

02:32:32.820 actually, but actually published a paper in this a while ago. So yeah, super interesting.

02:32:37.040 This is clearly top 10 questions I get asked all the time is, should I be on growth hormone?

02:32:42.340 My general answer is no, mostly through the lens of, I'm not convinced by any data that growth hormone

02:32:49.100 will promote longevity. But on the flip side, I've also softened my tone. I used to just think on first

02:32:55.840 principles, it made no sense because of if nothing else, the cancer risk. But of course, the other thing

02:33:01.400 that's odd is there's so many people taking growth hormone, especially athletes. And if there is a

02:33:08.320 negative consequence of it, it's probably not as big as I would have guessed. And, or it takes a lot

02:33:14.280 longer to show up than I would have guessed because we just don't see the trail of body bags that you

02:33:19.360 would expect.

02:33:20.280 I mean, that's the thing is you have Barry Bonds at 25 or 28 taking a bunch of growth hormone. What do you

02:33:24.000 expect? So maybe his risk of having an MI is increased twofold or threefold.

02:33:28.260 Would that be enough to see? Don't you think? Wouldn't a twofold increase show up?

02:33:32.540 In a non-randomized cohort of athletes taking an illegal drug? I don't think we have the data. I

02:33:37.960 don't feel confident. What I've learned in this work is that these hormones are regulated in ways

02:33:42.720 that we can't even begin to understand. I mean, it's so complicated. And so I'm nervous to go in

02:33:48.640 there and start perturbing it or replacing it if it's deficient. Because I just, I see all the

02:33:54.180 different things that can happen and how complicated that can be. So I'm a little bit more of a

02:33:58.640 minimalist, I guess, when it comes to these kinds of interventions. But my answer to my patients is

02:34:02.800 every line of evidence I can see about GH, at least, is that if you're worried, if you're

02:34:07.880 optimizing for longevity, it's hard to tell a story that that's going to help you. Now, if you're

02:34:12.660 optimizing for, I want to feel better. I want to increase my lean muscle mass. I want to decrease

02:34:16.880 my fat mass. I want to look better. Go for it if that's what you want. And so that's what I tell

02:34:22.080 people is if this is something that you're willing to understand that we have no evidence that it's

02:34:25.680 probably going to increase your lifespan, it may more likely even decrease it, but you like how it

02:34:29.920 makes you feel and look, then sure. So I would agree with all of that. I think the one area where I

02:34:35.380 would love to see more investigation, and this is certainly amenable to a clinical trial,

02:34:41.180 is does increasing IGF in patients with either very high risk for or early cognitive impairment

02:34:47.880 improve outcomes? I think the literature makes a plausible case that that could be the case,

02:34:52.920 and that strikes me as a subset in patients, at least for me clinically, that I am most interested

02:34:58.760 in knowing the answer to that question. If we take our high risk dementia patients, or worse yet,

02:35:04.300 patients who are already in the earliest stages of cognitive impairment, and take the ones who are at

02:35:10.380 lower levels, you know, the bottom quartile of IGF levels, will GH provide benefit?

02:35:16.780 Yeah, it's a great trial. What happens to your IGF one when you fast?

02:35:21.300 Oh, it plummets, and the numbers are kind of irrelevant because it's all a function of your

02:35:24.680 scale. So when I was on a ketogenic diet for three years with no fluctuation, my IGF level resided at

02:35:31.340 the 40th or 50th percentile for my age. So the lab that I use gives me not only my IGF level,

02:35:36.520 but in five-year increments, it gives me the distribution of it. So I can, you can very

02:35:40.700 accurately peg where you are. So for me, constantly being on a ketogenic diet put me at about the 40th

02:35:46.580 to 50th percentile for IGF level, which is not really surprising when you consider that a ketogenic

02:35:52.100 diet minimizes insulin. And also for most people actually keeps protein lower than you would

02:35:56.300 otherwise normally have it. So methionine plays such an important role in this, of course.

02:36:00.740 When I'm not on a ketogenic diet, which I'm not these days, except for in the peri-fasting period,

02:36:05.960 my IGF probably lives closer to the 70th percentile. On our scale, that's about 200.

02:36:12.240 When I do a fast, so I'll do a blood draw right before the fast. And then usually on day six or

02:36:18.180 day seven of a complete water only fast, that IGF will be in the eighties, nineties, which is more

02:36:25.540 than two standard deviations below the mean. What I find most interesting is how long it takes to

02:36:30.680 rebound. And it takes about six weeks to come back to normal. So, and have you ever, and measuring

02:36:36.540 growth hormone is one of the hardest things to do because it's so I've never measured it for that

02:36:39.780 reason. I just, I, yeah. I mean, it might be so high that you would be able to, to see something.

02:36:46.120 Let's talk about it. Cause I'd love to figure out the problem is I only fast in New York and ever in

02:36:50.180 San Francisco, but maybe I'll do a fast up here and swing by the lab and we'll do daily

02:36:54.220 or something. I guess we have to get an IRB these days. We can't do anything for fun,

02:36:58.140 but it's very interesting to see how much you can manipulate GH nutritionally.

02:37:04.080 Well, that's what, one of the things that we're really interested in understanding,

02:37:06.960 like in the future is understanding its role in this fasting feeding transition. And it goes up

02:37:11.240 in fasting, but as you say, IGF one levels go down. So you get this like selective GH resistance

02:37:17.360 in the liver, at least the question of is whether you're getting GH resistance in other tissues as

02:37:21.600 well. But why would you have this hormone go up and then lose its activity? It's kind of a funny

02:37:26.780 thing, right? And you see this in people with anorexia too.

02:37:29.440 Yeah. I've never actually understood it as well as I feel like I do now based on the way you've told

02:37:34.880 this story of, it's a great way to describe it, right? Selective GH resistance in the liver,

02:37:40.460 IGF is plummeting, GH is skyrocketing. Then the question, and we should know this and

02:37:44.980 eventually we will, what is GH activity in adipocytes? Is it actually maintained? And so

02:37:49.180 are you basically trying to mobilize fat as a way of avoiding starvation? Is it basically you're

02:37:54.980 trying to direct GH to act only on the periphery and not on the liver?

02:37:58.440 By the way, if you believe that periodic fasting improves longevity, and I certainly do,

02:38:04.240 Walter Longo does, we might come at it from different ways. It also begs the question,

02:38:08.720 is the benefit conferred by the reduction in IGF or the transient rise in GH?

02:38:13.340 Yeah. Or something completely different.

02:38:14.940 I never even really considered that the benefit could be conferred from the GH rise. Although that

02:38:19.780 seems less likely given that it's hard to imagine you, I mean, who knows though? Maybe,

02:38:24.320 maybe there's something, I mean, if biology teaches us anything, it's to stop acting like we know what

02:38:28.860 we're talking about. Exactly. Exactly.

02:38:31.560 Ethan, I want to go back to something because I know we've got to wrap up. You've got to get back

02:38:35.080 to stuff. You know, you talked about something at the beginning that I didn't know, which was you

02:38:40.560 went off to Vassar to get as far away from science as you could. And yet you had an experience there that

02:38:47.440 presented science in the right way. And I would argue the world is actually a better place for you

02:38:53.520 doing what you're doing now, which is not to say that you being the drummer of a band wouldn't

02:38:57.860 have been equally impactful, but maybe I'm biased. This is an interesting topic to me because

02:39:03.440 my biggest fear, or certainly one of my biggest fears in education is a retreat from science.

02:39:10.600 What could a parent think about out there as they think about their son or daughter? And I'm actually

02:39:16.720 even more afraid of it in the case of girls where there's this ridiculous, stupid, and I really think 1.00

02:39:21.740 it is stupid and incorrect that girls are just not going to be as good at science or math. 1.00

02:39:26.080 I think that most of that is actually a feed forward mechanism that's built into an expectation. 0.99

02:39:31.660 I mean, how do you think about this with your kids or how could I think about it with my kids

02:39:34.920 or anybody else thinking about this? Like, what can we do to make science as exciting as it can

02:39:40.780 possibly be? And if for no other reason, it's not, I even say this to people, it's not because

02:39:44.340 everyone has to become a scientist. It's an inability to understand science makes you a victim

02:39:49.180 for the rest of your life. Whether you want to admit it or not, if you can't at least have some

02:39:54.260 understanding of science and think critically, you are going to be overrun with propaganda and

02:39:58.980 nonsense. And the consequence of that is enormous. And we see it, you know, I don't think this fact

02:40:03.800 is necessarily correct today, but it was at the time, you know, that I knew it and it won't be

02:40:09.120 directionally off. At the time of the statistic, there's what, 565 members in the House of Representatives

02:40:16.040 in the Senate, 5 of them had degrees in science. So less than 1% of the lawmakers of this country

02:40:23.760 are trained scientifically. That just doesn't strike me as an appropriate balance. So what

02:40:29.980 happened at Vassar and how can we reproduce it? So those are two things that I think about a lot.

02:40:33.980 One is obviously we as a group of scientists need to do a better job of communicating what we do in an

02:40:38.700 interesting and exciting way to people without dumbing it down. And I think, honestly, I'm not just

02:40:42.720 saying this to shove sunshine up your ass, but I've had so many calls from people, either people I do 0.98

02:40:48.560 know or people I don't know who've said, I just heard this great podcast. And I think what you're 0.99

02:40:52.640 doing here is bold because you're not shying away from getting into the depth. It may be that a lot

02:40:58.120 of the conversation that we had around the science was over the head of the average layperson who hasn't

02:41:02.520 had science, but I still think it's valuable and they're going to go off and get literate about

02:41:06.600 something and they're finding it interesting. So I think we as a community need to do that. And whether it's you

02:41:10.440 or it's Ron Vale and iBiology or other things, which if you haven't seen it, I highly recommend

02:41:16.500 he's making these videos. In fact, they did one with Dave Sabatini, a series on mTOR.

02:41:20.680 Yeah, I saw that one.

02:41:21.500 Yeah. If we're going to get more people interested in science, we got to make it more interesting.

02:41:25.860 The second thing is there's this trend and you know this because you've got kids of hyper

02:41:30.520 specialization earlier and earlier in life. I could afford to go to medical school

02:41:35.400 making the decision as a sophomore in college. That's almost impossible to believe that that

02:41:41.460 could happen today. I mean, it could happen, but you'd have to take time off. But what I find with

02:41:44.940 our trainees is that people who don't, two things happen. One is people who get into science at a

02:41:50.680 young age get burned out and don't like it by the time they get to be, you know, a fellow or a resident

02:41:55.620 or attending. And the second thing is the people who don't get into science, like me,

02:41:59.560 the Ethan Weiss's who go to Vassar in 2019 and not 1991 find it way too much of an obstacle to get

02:42:08.580 in now. Like it's just too complicated and too many barriers. So one of the things that I'm

02:42:12.740 passionate about and I'm actually working on it, I can't really talk about it now, but we're going

02:42:16.680 to talk about it soon. We're working on some ways to help MD only medical students and beyond

02:42:22.220 get exposed to like real intensive two and three year experiences in the lab. So the kind of thing that

02:42:28.380 used to happen routinely where people would go and work at the NIH like you did or other things

02:42:33.680 where they can get exposed to science in a meaningful way, even if their career is not

02:42:36.780 going to be as a PI running a lab, it will help increase the scientific literacy, having, being

02:42:41.900 able to speak both these languages. So that's something I'm super passionate about.

02:42:45.180 Because traditional medical school is actually not scientific training.

02:42:48.260 I mean, it's gone now. So, you know, I've heard this saying, I don't know who this one gets

02:42:51.380 attributed to, but the gist of it is that we've taken the science out of medical school and the

02:42:54.720 medicine out of graduate school, right? So our graduate students are

02:42:58.240 craving more understanding of physiology and medicine. Our medical students are not really

02:43:02.180 missing what they're not getting, but they're not learning science anymore. And so once that

02:43:06.340 happens, then the likelihood that that student who's not done a PhD, who's never worked in

02:43:10.740 a lab is going to have the courage or the ability to go work in a lab later on in life is close

02:43:16.620 to zero. And the thing is, what was great about the way it happened for me was that I came

02:43:20.540 to it without much preconceived notion and dogma. So I came to it with just native curiosity

02:43:25.920 and sort of like, how should things work? And look, I wish my basis of scientific knowledge

02:43:32.940 was, was greater than it is. I I'm jealous of people like, you know, my MD PhD friends

02:43:37.560 from Hopkins, like Dave Sabatine, those guys, I'm totally jealous of how deep their knowledge

02:43:41.960 is, but I have a different perspective and I bring the clinical work that I do and the

02:43:46.560 things that I think about when I'm seeing patients and have thought about. So it's a different

02:43:50.200 way.

02:43:50.420 We didn't get a chance to talk about it, but you've been really, I mean, as though everything

02:43:55.200 we've described, your entire clinical career, your amazing stuff you're doing in the lab

02:43:59.620 isn't enough to keep you busy. And oh, by the way, you also have a family. You decided

02:44:04.400 like, I mean, I feel like you did it in a week. Cause I remember when you called me and told

02:44:08.200 me about it. And then a week later you had a prototype and then two weeks later you had

02:44:11.840 your next brother. You started a company last year.

02:44:14.280 Yeah.

02:44:14.940 So can you tell us a little bit about that? Cause half my patients now are using this device.

02:44:18.860 We could do an entire podcast on this and maybe I'll have a podcast and bring you on. Cause

02:44:23.520 I do think there's some of this stuff. Like when I first called you, I remember calling

02:44:26.240 you, it was like a Saturday or Sunday and you were generous to take the time to help

02:44:30.020 me think about this. And we talked a lot about the behavioral science, which, and the behavioral

02:44:33.740 economics, which is something we've talked about a lot at our time on the Virta advisory

02:44:37.460 board. And I think it's, to me, it's the most interesting part of this whole obesity

02:44:41.020 story, right? Is that we're talking about a behavior at its core and a change in behavior.

02:44:46.700 And how do we, at the absolute most basic level, how do we use technology to help people

02:44:53.020 change their behavior? And that's something that I've been thinking about for a long time.

02:44:56.820 And to make a really, really, really long story short, I got together with an amazing

02:45:01.880 team of people who'd been doing work in weight loss. And actually when I told them what I wanted

02:45:07.100 to do, they both told me to get lost. They said, they're never going to do a weight loss

02:45:10.180 startup again. They had started and sold a company to Weight Watchers and said, it's too hard.

02:45:14.580 People say they want to lose weight, but they actually don't. They don't have the tools to

02:45:17.960 do it. So short version of a long story, we thought, well, gosh, you know, one thing that

02:45:23.720 is true, I think about weight loss in general for people who struggle is that we're not giving

02:45:29.300 them the information they need to be able to make the change in behavior. The only information

02:45:33.840 that we give people is the information they get from stepping on their scale. That's an incredibly

02:45:37.800 lagging indicator. I mean, you could have an entire gallon of ice cream tonight at 11 o'clock

02:45:42.060 and step on the scale tomorrow and you may not have any difference in your weight. So

02:45:45.000 it's relatively useless on a day-to-day level. And by the time you actually get information,

02:45:49.500 it's not actionable. I would add one more thing, which is the amount of movement in water

02:45:54.700 alone makes it almost impossible. For example, the average woman who is still experiencing a

02:46:01.160 menstrual cycle can easily fluctuate by eight pounds in a month. It's a four kilogram average

02:46:07.360 fluctuation in water weight. So can you imagine if a woman says, I want to lose 10 pounds and she's 1.00

02:46:14.340 using her scale, it's just impossible. And you can build in moving averages and other things,

02:46:20.200 but again, you're not giving them the information they need to be able to make the changes if they

02:46:23.400 want to make a change. So what's interesting and unique about it. So when I came to the Virta

02:46:27.680 advisory board, I'd never done keto before. And frankly, I kind of thought it was a little bit

02:46:31.580 weird. I don't know if you remember that first advisory board meeting we had at the Mexican 0.99

02:46:35.020 restaurant down there. And I still joke about this. In fact, I think I've joked about it with

02:46:38.240 Sami and Steve that I show up at this thing and like there's bowls of guacamole there with spoons

02:46:42.720 in them and like, you know, fajita dishes without wraps. And like, there was no, not a carbohydrate

02:46:48.640 to be found on the table. I thought this is like a cult. Like this is, this is weird. And I never

02:46:53.240 really had any curiosity about doing it. And then I tried it myself and I realized that the one unique

02:46:58.140 fact about this diet that's different from all other diets that I know of other than fasting for a

02:47:02.520 week is it allows you to follow biomarker. And that if you use that biomarker as a guide to help

02:47:08.000 your own adherence, it can help reinforce some of these behaviors. And maybe that would be sufficient.

02:47:14.320 So we finally decided to try this thing. And one of the issues, as you know better than anybody,

02:47:19.560 testing your ketone level, whether it's through blood or breath or urine, has all kinds of issues.

02:47:24.720 And there's imperfection, but we figured if we could build a good sensor, I'm not talking about like a

02:47:29.840 lab level sensor, but a good sensor that people could use, they could carry around that would be

02:47:34.580 portable and accurate enough to help guide these behavioral changes that we could use that as the

02:47:41.340 basis of basically replacing a lot of the human intervention that goes into programs like Virta,

02:47:47.360 right? So Virta is successful at what they do in a different population, right? The population of

02:47:52.420 people not really primarily trying to lose weight, but more trying to manage their diabetes,

02:47:56.040 but that takes a lot of human capital. So could this information help replace that and kind of make

02:48:02.840 it more economically feasible? And so basically through just a series of incredibly fortunate

02:48:09.000 accidents and beyond hard, hard work from this team, we were able to get the sensor, swap out a

02:48:14.820 alcohol sensor for an acetone specific sensor and got this thing to market quickly. It didn't feel like

02:48:20.540 a day, but it was fast. I mean, we, we basically, I think you and I spoke in August.

02:48:24.580 I mean, I feel like I had a prototype in October because I managed my calendar through my fasts,

02:48:30.800 which are always beginning and quarter. And then of course, by the, by December had the,

02:48:35.020 the version I have now, which is great. And that's the one a number of my patients have.

02:48:39.000 I think it was November. And so, yeah, we are doing this kind of slow.

02:48:41.840 So tell people, I don't even think we've mentioned the name of the company. So what's the name of the

02:48:44.680 company?

02:48:45.080 It's called, well, it's spelled K-E-Y-T-O and most people pronounce it keto as you would pronounce

02:48:50.660 ketogenic diet, but we were sort of intentional in being vague about whether you could call it

02:48:55.200 key two, if you wanted to, and it leaves it open to the idea that it could be more than just

02:48:59.320 ketogenic diet. And so the company is organized around one is giving people the information they

02:49:04.760 need through this sensor, which, you know, links up to an app to giving people information they need

02:49:08.800 to be able to succeed at whatever they're doing. So you know how to do keto. You've done Ted talks on

02:49:13.620 keto, but an average person hears about ketogenic diet and they're like, well, I don't know how to do

02:49:17.700 that. So we give people the information they need to be able to do the diet. And then at some

02:49:21.620 point, and it won't be now, but at some point we're going to help people get to, you know what

02:49:25.380 to do. You're measuring how well you're doing, but how you actually, how can you enable that?

02:49:30.060 How can you get people the food? So I cook 90% of the meals. I've got my lunch that I brought with

02:49:35.120 me over there in a bag, but a lot of people don't have that luxury. And so they'll go out here,

02:49:39.740 this food court and want to get lunch. And so can we help deliver that food to them at some point?

02:49:44.600 So that's sort of the big picture of what we're doing. And it's been incredibly fun. I mean,

02:49:48.720 I've been at UCSF now, I've been in this office for almost 10 years. And I, I do think I'm like

02:49:55.660 one of the luckiest people in the world to be able to do the things I do both combining this clinical

02:50:00.020 work, which is just a gift with this lab work, which is basically like a game. It's like a glorified

02:50:05.560 living, breathing video game that we get to try and solve puzzles every day. And I get paid to do it.

02:50:10.260 But this thing has been fun to do and it happens at a different pace. As you said,

02:50:14.740 it's like a normal cycle for me in the lab is like five years. This thing happened in three months.

02:50:19.240 And so it's fun to kind of exercise different muscles and think about business for the first

02:50:24.360 time in my life. I mean, I haven't really thought about business before and it's fun. I actually enjoy

02:50:28.060 it. I think about, I love thinking about the human psychology and, and the behavior. So it's been

02:50:32.420 kind of a fun adventure. We'll see how it goes.

02:50:33.980 Awesome. And you've got one for me to take back to one of my, I'm seeing one of my patients next week who

02:50:38.160 I was with the other day and I was talking about it with him. Like I had already given him one and

02:50:43.300 he's like, well, what are you talking about? And I was like, well, I'm going to see Ethan. So I'll

02:50:46.760 get you one. You're someone that I follow on Twitter. Cause I always enjoy, I mean, you know,

02:50:50.680 for me, Twitter is just a great way to curate reading and stuff. And I love following your stuff.

02:50:56.080 So where can people find you?

02:50:57.700 That's probably my biggest place that I spend time social media wise. And my Twitter handles

02:51:02.580 at Ethan J wise, I have to say, I probably had a significant problem and my wife would

02:51:07.500 say that I had a more than significant problem with the amount of time I spent over on there.

02:51:10.960 I'm not there as much anymore just cause I can't keep up that and my job and my family and this

02:51:16.980 startup thing. So, but I do try to respond, you know, somebody sends me a message. I'll try to

02:51:21.340 respond. Even if once a week, there's a great study that you're highlighting. I find that super helpful.

02:51:26.200 Yeah. It's fun. Look at Twitter. I actually went down there and visited a friend of mine as the CFO at

02:51:30.880 the company, Ned Siegel. And he invited me to come. He said, you're so active, like you're a power user

02:51:34.600 in science and medicine. You should come visit. And it was really sweet. He brought, he invited

02:51:37.860 me to come down and we toured through the facility. I'd never been there before, but I think they have

02:51:40.960 something going there that talk about increasing scientific literacy. It's, it's, and I know I've

02:51:46.080 my own patients who follow me and they tell me they learn a lot. They learn a lot about medicine

02:51:50.480 and science. They learn a lot from you. And you don't have to be a stranger to, unless you've been

02:51:54.120 in the Himalayas hunting Yeti for the past five years, you certainly realize that right now Twitter

02:51:58.100 is under attack. And I feel bad every time Jack Dorsey has to go on a podcast, it's just basically to get

02:52:03.380 skewered. I understand why people are upset about the role of Twitter and politics and fake news and

02:52:07.600 stuff. And I think that's a super upsetting problem. Fortunately, in the world that we play

02:52:12.440 in Twitter, I still find it more positive than negative. In the end, I've got a bunch of people

02:52:18.240 I follow who basically curate some of the research I just wouldn't be able to keep up with. Cause I'm,

02:52:23.140 you know, I'm really narrow in what I do, but I need to know what's going on in these fields that

02:52:28.140 are adjacent. And that's a great way to do it. So Ethan, I can't thank you enough. This has been

02:52:32.080 awesome. And I suspect this won't be the last time we have to sit down and have a formal discussion.

02:52:37.640 And we'll obviously have many more informal ones. Awesome, Peter. Thank you so much for coming up.

02:52:43.920 You can find all of this information and more at peteratiamd.com forward slash podcast.

02:52:49.160 There you'll find the show notes, readings, and links related to this episode. You can also find my blog

02:52:54.860 at peteratiamd.com. Maybe the simplest thing to do is to sign up for my subjectively non-lame once a

02:53:00.740 week email, where I'll update you on what I've been up to the most interesting papers I've read

02:53:04.920 and all things related to longevity, science, performance, sleep, et cetera. On social,

02:53:10.280 you can find me on Twitter, Instagram, and Facebook all with the ID peteratiamd. But usually

02:53:15.520 Twitter is the best way to reach me to share your questions and comments. Now for the obligatory

02:53:19.840 disclaimer, this podcast is for general informational purposes only and does not constitute the practice

02:53:24.540 of medicine, nursing, or other professional healthcare services, including the giving of medical advice.

02:53:30.140 And note, no doctor-patient relationship is formed. The use of this information and the

02:53:35.640 materials linked to the podcast is at the user's own risk. The content of this podcast is not intended

02:53:40.620 to be a substitute for professional medical advice, diagnoses, or treatment. Users should not disregard

02:53:46.140 or delay in obtaining medical advice for any medical condition they have and should seek the

02:53:50.580 assistance of their healthcare professionals for any such conditions. Lastly, and perhaps most importantly,

02:53:56.360 I take conflicts of interest very seriously. For all of my disclosures, the companies I invest in and

02:54:01.720 or advise, please visit peteratiamd.com forward slash about.

The Peter Attia Drive - May 06, 2019

#52 - Ethan Weiss, M.D.： A masterclass in cardiovascular disease and growth hormone - two topics that are surprising interrelated

Episode Stats

Summary

Transcript