EP22: Dr. Philip LeBoit on being curious, tips on diagnosis and avoiding error, and fallibility

***Please excuse my voice - I recorded this towards the end of a several weeks-long illness!***
What is the difference between late medieval and early Renaissance Italian painting? Dr. Philip LeBoit knows! (find out how!) (I still don't know the actual answer....) Dr. Philip LeBoit is Professor of Pathology and Dermatology as well as Division Chief of the Dermatopathology Service at the University of California, San Francisco. He founded the UCSF Dermatopathology and Oral Pathology Service in 1987. He trained in Anatomic Pathology at University of California, San Francisco and then in Dermatopathology at New York University under Dr. Bernard Ackerman and at New York Hospital-Cornell Medical Center under Dr. N. Scott McNutt. He has written several academic textbooks, including one of my favorites with Dr. Guido Massi, Histologic Diagnosis of Nevi and Melanoma. He was editor-in-chief of the American Journal of Dermatopathology from 1997-2006 and an associate editor of the Journal of Investigative Dermatology.

Transcript

[00:00:00] Christine Ko: Welcome back to SEE HEAR FEEL. Today, I have the honor speaking with Dr. Philip LeBoit. Dr. Philip LeBoit is Professor of Pathology and Dermatology as well as Division Chief of the Dermatopathology Service at the University of California, San Francisco. He founded the UCSF dermatopathology and oral pathology service in 1987. He trained in anatomic pathology at University of California, San Francisco, and then in dermatopathology at New York University under Dr. Bernard Ackerman and at New York Hospital Cornell Medical Center under Dr. N. Scott McNutt. He has written several academic textbooks, including one of my favorites with Dr. Guido Massi, Histologic Diagnosis of Nevi and Melanoma. He was Editor-in-Chief of the American Journal of Dermatopathology from 1997 to 2006, and an Associate Editor of the Journal of Investigative Dermatology. Welcome to Phil, and thank you for spending time with me. 

[00:00:51] Philip LeBoit: Oh, it's such a great pleasure and thanks for, thanks for having me on. 

[00:00:58] Christine Ko: One of my first questions is - if you wouldn't mind sharing something personal about yourself with me and the audience, something to humanize you since you're larger than life...

[00:01:06] Philip LeBoit: Okay. So amongst the things that I'm totally incompetent at are dancing. One of the most embarrassing moments of my life was at a meeting of the International Society of Dermatopathology in Brazil. If you've ever been to Brazil, much of Brazilian culture involves dancing, and medical meetings in Brazil are really an excuse to have a dance party. After dinner, the dance floor was empty. The orchestra had just started up and a very beautiful woman, who I had no idea who she was, walked over to my table and picked me, picked up my hand, and led me onto the dance floor in front of about 300 people. I can only surmise that she was on a mission to embarrass me. I found out afterwards that she was a recently minted Brazilian dermatopathologist. 

[00:02:01] Christine Ko: Oh, that's funny. That's a nice story. All right. Well, so with that introduction, I'll go into one of my first questions, which is, I wanted to hear how you define and use emotional intelligence.

[00:02:16] Philip LeBoit: Okay. I think the key to emotional intelligence is to try to think about what somebody else may be feeling and what the possible range of intents behind a statement is instead of jumping to the conclusion that a statement means a certain, very specific emotional intent. It's very easy to assume that if somebody said something, they must mean something critical or they're angry, et cetera, but it may have absolutely nothing to do with that and it may have nothing to do with you. I think a key thing is realizing what the range of human emotions is, in not jumping to conclusions, and being curious. 

[00:03:01] Christine Ko: Being curious is a really important thing because it can help us not just assume things. You're widely recognized as a very accomplished dermatopathologist and pathologist. I wanted to know if there's a way that you currently, or in the past, hone your visual perceptive skills, and how much you think being a good pathologist or dermatopathologist is talent versus hard work?

[00:03:28] Philip LeBoit: I don't think the innate talent that you need to bring to the game is that great. There definitely are people who are to visual interpretation what I am to dancing. You can show them a rectangle and a square, a hundred times in a row, and time number 101 they still won't be able to tell you, which is the rectangle, which is the square. But if you have some very basic skills, then I think it comes down to practice, practice, practice, or Malcolm Gladwell's 10,000 hours. In terms of honing those skills, there's a variety of ways to do it. I have to pay tribute to my father in this respect, who was an artist, and he dragged me to most of the major art museums in the world when I was a child and talked about the very subtle differences between late medieval, Italian painting and early Renaissance, Italian painting; and things like that. So I grew up looking at things with a very critical eye. 

[00:04:28] Christine Ko: Yeah, that's interesting that you have that background because Almut Boer said that, as a child, I think one or both of her parents were artists. And so there were art books at home and she learned about foreground, background; or just concepts that are very well known in art and the art world for anyone who's familiar with art, but things that we don't really commonly talk about so much when we are thinking about making a diagnosis on a slide. Do you have a diagnostic process? 

[00:05:00] Philip LeBoit: Bernie Ackerman was very fond of saying that he had criteria for everything. And indeed he did, but he also recognized the fact that - and he would say this - when you see your friend's face, you don't say, what are the criteria for recognizing this as my friend, Dan, for instance. You just say that's Dan. 

[00:05:26] Christine Ko: Yeah. 

[00:05:26] Philip LeBoit: With many of the diagnoses we make, ones that pathologists call instant pattern recognition: dermatofibroma, basal cell carcinoma, et cetera, (I'm talking about the 95% of them that are completely characteristic), it's similar to that. It's a little more difficult to identify what my process is when I'm stumped, which happens a fair deal. That's when one starts to actually try to look at things through the lens of criteria. But I can also recognize that that's sometimes very deceptive. I think anybody's practice foil has the experience of looking at something at 6 in the evening, and then looking at it again the next morning with fresh eyes and seeing something that looks very different, and how you would weight the criteria might be very different depending on what time of day it was. Things look more threatening at 6:00 PM. 

[00:06:29] Christine Ko: I've been wondering about steps to reduce diagnostic error, to reduce the times when you actually did miss something that ideally you would've seen. I don't know if there are tips that you've developed, or ways you've developed over the years to try to prevent that kind of not seeing something important.

[00:06:48] Philip LeBoit: Starting with very basic things. One thing I learned from Dr. Ackerman was, just simply hold the slide up before putting it under the microscope. And it's ironic that one of the cases, that his perhaps most famous miss of all times, the bisected polypoid nevus that ended up as the "case in Philadelphia" - it was a malpractice case against Dr. Ackerman, a bisected melanocytic lesion where one section had nevus, the other section had melanoma - and he only looked at one of the two. 

[00:07:21] Philip LeBoit: So starting with that, and just making sure that you're looking at every single piece of tissue. In terms of avoiding errors - I'm just going to talk about melanocytic lesions for a second – one thing is to look for breaks in pattern. It's something that I think is under emphasized. For instance, if you are looking at a melanoma in situ, and there's not been a previous biopsy, and the patient has severe solar elastosis, if there's an area of the dermis that's pink, look for subtle involvement by desmoplastic melanoma. If you are looking at a melanocytic lesion and trying to figure out if it's melanoma in situ or not, look at the rete ridge pattern. If there's a change in the pattern, if it goes from elongated ridges to a flat pattern in some areas, that could be a clue to melanoma in situ. I've also trained my eyes to look along the surface of a cornified layer because sometimes, as we all know, pagetoid scatter can be really subtle. And it depends, you know, some days our lab comes out with really robustly stain slides and some days they're faint, and some days, it may be hard to see melanocytes against a background of keratinocytes, but you'll sometimes see individual melanocytes in the cornified layer when you haven't realized that there's pagetoid scatter.

[00:08:54] Philip LeBoit: Those are all things in which I've been burned. I've gone back to the cases that I thought were dysplastic nevi and seen the scattered melanocytes in the cornified layer, even without seeing the pagetoid scatter. And I've gone back to the areas of pink dermis that had incredibly subtle desmoplastic melanoma. Getting back to the foreground background thing: there was a study done - Joseph Smolle, I think, at University of Graz in the 90’s, where they did keratin stains on melanocytic lesions, and just looked at the shape of the epidermis, and then trained a computer on it. And the computer program could recognize melanoma with 80% accuracy just based on the irregularity of the rete ridge pattern.

[00:09:49] Christine Ko: I also wanted to ask you about error.

[00:09:51] Philip LeBoit: I think there's an idea that we're supposed to be infallible, especially if you're at a level where you're getting consultations from other people, having an aura of infallibility is sort of part of the game that everybody plays. And it's just not true that experts are infallible. And I want to be honest about it. 

[00:10:17] Christine Ko: How is it that you are able to be honest about that when, you know, the prevailing medical culture kind of is really against that from when we're, I think, students and trainees? 

[00:10:32] Philip LeBoit [laughs]: yeah...

[00:10:32] Christine Ko: That's, it's just your personality? 

[00:10:36] Philip LeBoit: I guess so. I'm kind of oblivious to the consequences of it. You know, if somebody says, you know, oh, that guy's an idiot, I'm never going to send another case his way, you know, fine. That's, you know, completely fine with me. I get to go home earlier.

[00:10:52] Christine Ko: Everyone makes mistakes. We're all going to miss something at some point. 

[00:10:56] Philip LeBoit: Right, right. 

[00:10:57] Christine Ko: None of us is perfect. It is really nice that you help promote the idea that none of us is infallible. It means a lot that you do that given how respected you are. Is there a lesson that you wished you had learned earlier? 

[00:11:11] Philip LeBoit: Yeah. One was don't follow leaders watch the parking meters, which was a line from a Bob Dylan song of 1965 or 6. I think it's very easy to, especially if you were a young person in the field to find charismatic, older people and swallow everything that they say. And you have to think about the history of medicine and how much of what was taught in 1930 is ludicrous by today's standards, and really recognize that, you know, hopefully a smaller percentage, because I think the scientific method has become more pervasive and opinion is less pervasive, but lots of the things we say and teach are going to be revised in the future. So I would say no matter how much you admire somebody, take what they say with a grain of salt. 

[00:12:02] Christine Ko: That's good advice. Yes. Do you have any final thoughts?

[00:12:10] Philip LeBoit [laughs]: That sounds a little bit like, like an epitaph.

[00:12:15] I do have a final thought and it's mostly for the younger people who might be listening to this podcast. So we have AI, we have genomics... there's so many interesting things to be done still in the field of dermatopathology that these new technologies have opened up that need human intelligence and creativity to apply and push the field forward. So in those respects, I think it's an absolutely great time to be a dermatopathologist. 

[00:12:45] Christine Ko: Well, thank you so much for spending time with me.

[00:12:48] Philip LeBoit: Well, it's really been a pleasure, Christine.