#020 - Dr. Erik Jensen - Bronchopulmonary dysplasia research

Dr. Erik Jensen is an attending neonatologist in the division of neonatology at the Children's hospital of Philadelphia. He is an assistant professor of pediatrics at the Perelman School of Medicine at the University of Pennsylvania.

Dr. Jensen’s research seeks to improve the long-term respiratory health of premature infants through: (1) clinical and translational studies that aim to develop evidence-based strategies to prevent and treat BPD; (2) novel characterization of disease severity and phenotypes in BPD; and (3) application of health services research techniques to investigate the association between hospital-level factors and neonatal morbidity and mortality.

Dr. Jensen’s research is supported by grant funding from the National Heart Lung and Blood Institute (NHLBI), the National Institute of Child Health and Human Development (NICHD), and the American Lung Association (ALA). He is a member of the International BPD Collaborative pharmacology working group and the International Neonatal Consortium (INC) BPD working group. He has received multiple clinical and research awards. 

_____________________________________________________________________________________________________

As always, feel free to send us questions, comments or suggestions to our email: nicupodcast@gmail.com. You can also contact the show through instagram or twitter, @nicupodcast. Or contact Ben and Daphna directly via their twitter profiles: @drnicu and @doctordaphnamd. 

enjoy!

As always, feel free to send us questions, comments, or suggestions to our email: nicupodcast@gmail.com. You can also contact the show through Instagram or Twitter, @nicupodcast. Or contact Ben and Daphna directly via their Twitter profiles: @drnicu and @doctordaphnamd. The papers discussed in today's episode are listed and timestamped on the webpage linked below.

Enjoy!

Transcript

Ben: 0:41

Hello, everybody. Welcome back to the show. Daphna. How are you how was your call last night?

Daphna: 0:47

It was tolerable and I'm feeling energized by you know, working on the podcast every time. That's right.

Ben: 0:53

That's right. We're very excited about our guest today. Today we have the honor and the pleasure of having on the show Dr. Eric Jensen. Dr. Jensen is an attending neonatologist in the Division of neonatology at the Children's Hospital of Philadelphia. He's an Assistant Professor of Pediatrics at the Perelman School of Medicine at the University of Pennsylvania. Dr. Jensen's research seeks to improve the long term respiratory health of premature infants through one clinical and translational studies that aim to develop evidence based strategies to prevent and treat BPD to novel characterization of disease severity and phenotypes and BPD and three, application of health services research techniques to investigate the association between hospital level factors and neonatal morbidity and mortality. Dr. Jensen's research is supported by grant funding from the National Heart, Lung and Blood Institute, the National Institute of Child Health and Human Development, and the American Lung Association. He is a member of the International BTG collaborative pharmacology Working Group and the International neonatal consortium BPD working group. He has received multiple clinical and research awards. Dr. Jensen, thank you so much for being on the show with us today.

Erik Jensen: 1:57

Thank you for inviting me. It's a pleasure to be here.

Ben: 2:00

Thank you. Well, I guess we wanted to start off this interview by asking you really? What What were the driving factors that led you to pursue a career in medicine and specifically in neonatology?

Unknown: 2:14

Yeah, so I think I always knew at least from an early age, I thought I wanted to be a doctor, I don't know that there was a salient experience that sent me in this direction. You know, when I read applications to a fellowship program, that you often will see someone say, Well, I had a sick relative, or I had this experience where I was in the hospital, and someone was nice to me, or they weren't nice to me, and so on, and so forth. And I never had anything like that. I thought medicine was interesting. I was interested in science, and I liked working with people. And so that led me into medicine. When I started in medical school, I think like many of my classmates, I was convinced I was going to become a surgeon, because I wanted to work with my hands and surgeries. Cool. And it's, you know, the fancier profession. And I had not thought that pediatrics would be of interest to me, but I liked kids. I liked parents actually, which most people say that's why they don't go into pediatrics. But I found parents to be very easy to work with, for the most part have, in general motivated to help their kids get better. whereas adults, I think, are less motivated to make themselves get better in many instances. So I pursued pediatrics. And similar to sort of wanting to be a surgeon, I was convinced that I was going to be an interventional cardiologist when I got started, because that was sort of the cool version of Pediatrics where you still got to do procedures and things. So I went to the cardiology program, were at my residency, and I said, Hey, I want to do research. Can I do a project and no one ever got back to me? And I followed up again, I never heard back and I don't know how many times I tried, but I tried a couple and and there wasn't a lot of interest in in helping me out and a number of people from my residency went off to do cardiology. So I think that there was a whole bunch of people and, and so, you know, dedicating a lot of effort to one resident

Ben: 4:21

and typical cardiology fashion, right? Yeah, maybe

Unknown: 4:24

it wasn't all that necessary. So I had just done it was in the middle of doing my first neonatology rotation as a second year as a first year resident early on in training anyway, and I enjoyed it. I thought that the ICU was was cool and that the people I worked with were great. And so I went to rich Polin, who was at the time the head of neonatology at Columbia and is still you know, like a huge investigator and person in the field. And he immediately expressed an interest in helping me find a project and I met with him like a month later, and he had a few ideas in mind and all the people that were in charge of this project had agreed to work with me if I wanted to. And so I picked one and we did a project, I got to present the PHS. And so it was great that I had found, I think an area of medicine that I liked doing and area of medicine that they had sort of welcomed me and wanted to work with me and, and so that that felt good. And I did my second rotation, you know, as a more mature resident, and I enjoyed it even more and just sort of went from there and I applied for fellowship, I did a house duck year, like a hospitalist year in the NICU before leaving New York to come to Philly. So I really liked the that aspect of medicine, that aspect of pediatrics and I was like shunned by the cardiologist. So I and was welcomed by the neonatologist. So that's a good way to

Daphna: 5:57

Yeah, who knows what you'd be doing? Just answered your emails, right?

Ben: 6:02

I'll probably pulmonary hypertension research, I guess

Unknown: 6:05

it could be Yeah, I mean, who knows? Right? It's, it's, it's one of those things where you wonder. So New York was my I probably shouldn't admit to these things. But Columbia was, was very high on my rank list, but not my first choice of residency. So and I'm very happy that I ended up there and living in New York was amazing. But you know, you, as you look back on your career, you certainly wonder, well, what happens if I had matched in this other place that I hadn't had the same mentors like now? Would I be a cardiologist? And then what I not live in fairly when I not be married to my wife would not have my two kids like, yeah, and then you get down this rabbit hole. And I'm just gonna stop thinking about all that. Just be grateful that I am where I am.

Daphna: 6:50

That's right. So true. So true.

Ben: 6:54

I was going to ask then, since we're talking about what if you did pursue a Master of Science in Clinical Epidemiology. And and I think one of the one of the things that exemplify your work is this very elegant and rigorous scientific methodology. And so I'm curious to hear your thoughts on what your thoughts are on this master of science that you pursued? Has it helped you in being the researcher you are today? And did you expect that this was going to be something that would be so valuable when you did pursue that,

Unknown: 7:24

that track? Yeah, I knew coming out of residency that I wanted to obtain more research, training, more clinical research training. But I was not all that familiar with the MSC II that masters of science and clinical epidemiology at Penn that I, that I had, I, I knew what mph was, I knew that was sort of a degree in clinical research or something along those lines. And I knew kind of that pen didn't offer an MPH, but they offered this MSC e thing as an alternative. And when I interviewed him for fellowship, they said, Yeah, if you want to do clinical research, and you want to sort of get a degree in that this is the degree you get, and so on and so forth. So I had a general sense that that was the direction I was supposed to go down, Penn has done a very nice job of making it possible for trainees to obtain that training at no cost to us through grant funding. And so I was able to apply for that and get one of those. And I'm incredibly grateful that I had that chance, because that the combination of that grant, or that training, that master's degree and the coursework and whatnot that went with it, and the mentors that I had in my division, I think had been the most transformative education that I've had, in my, you know, 20 years from grade school on B, cause it really provided the sort of nuts and bolts of how you do research how you do research. Well, how you do statistics to do them. You know, I think that in in medicine, like in medical school, and even in residency, you do so much that you end up forgetting or not using and I think it builds a foundation, right? But you don't like I don't remember anything from my rheumatology rotation, and so on and so forth. Right. But this was stuff that like from day one till graduation, every single thing felt practical, and I'd say probably 80% of what I learned, I've applied in some way to my research career and then taking those tools and sort of been going to my mentors with projects or having them provide project ideas to me and sort of thinking through how to do them and then critiquing my work to be combined with the research training has has been exceptional and I am not alone. In this opinion. I think the people that I work with here at Penn feel the exact same way about about that sort of training and mentorship that they've had as well.

Ben: 10:03

So you would say more useful than memorizing the Krebs cycle.

Unknown: 10:07

Because I've forgotten it. And that's about it.

Daphna: 10:14

You know, I've always thought about that when I was going through my fellowship. And admittedly, I did not have a strong research background. And And thankfully, I had mentors that really helped with that, you know, unlike other doctoral programs, say psychology or other science fields, that statistical training, that research methodology is just a part of the, the program. And so it's it's so interesting that in medicine, we don't do that even with, you know, ongoing training, like a fellowship. So I wonder if more and more doctors are going to be doing, you know, doctors who want to stay in research doing kind of what you did, and hopefully there'll be more opportunities for that.

Unknown: 11:01

Yeah, I think people of my age, you see more and more that like, when you look at articles that are published, right, many more people now than before, don't just have an MD they have something else, and maths or an MPH or a PhD. Whereas the many of the folks who are big name investigators in my mentors generation, most of them sort of just had on the job training. The MSC that I had, would really not have been available to them. It just happened that my mentors had an MSc because they trained at McMaster and McMaster was one of the first institutions to develop that program. It's really the institution that was if not the most important one of the most important in developing sort of the concept of evidence based medicine and clinical trials and sort of how we develop it answer in a robust way clinical questions. So I think they had that training as an advantage to them because of where they worked. And I've now had my training as an advantage of where I work. But more and more I think young people are obtaining that training, if they end up wanting to do clinical investigation.

Ben: 12:20

So let's, let's jump into a little bit of of your work on BPD. And I wanted to talk to you about obviously, this 2019 paper where you lay out sort of a new classification, you and the rest of the group from the neonatal Research Network. But obviously you're you're the first author on this paper, where you lay out sort of the some criteria, new criteria to define BPD at 36 weeks and something that were most of us now refer as the Jensen criteria. My first question is how many years into practice where you went up paper came out?

Unknown: 12:54

Well, I guess it came out in 2019. And I graduated fellowship, I think in 2014. So it would have been five years. I think that's right. Yeah. Five years.

Ben: 13:05

And so my my so then my follow up to questions are it's a very ambitious project, right? I mean, for somebody who is a young, a young attending to say, I'm going to tackle the BPD definitions with this with the neonatal Research Network. And did you when you did take on that project? Did you ever think that the this paper would be as impactful as it has become?

Unknown: 13:33

I don't know. I don't think so. No? Well, the motivation for the paper came. I had I attended the, the, I think it was in 2018, or late 2017, when the NI CHD held a workshop on BPD. And they published the results of that in Journal of Pediatrics in 2018, I think. And I had received an email that was forwarded me by the fellowship director at CHOP, saying, Hey, you and Nick Bateman who's one of my colleagues should think about going to this, it seems interesting. So we went. And I think they were like a little bit surprised to see these young people there. But they were very welcoming. So we appreciate it. But right, you're sitting in a room with all these famous people, and they're sort of looking at you like these kids. But they were very nice to us. And so we appreciated the opportunity to attend and sort of rub shoulders with the Giants. But one of the groups that both Nick and I joined at the meeting was focused on, you know, how should we define BPD and so that in the last sort of big severity based definition, right, the joven Baccalaureate paper from 2001, and then previously in the 1970s, were that sort of two big BPD workshops, all sponsored by the NIH, and they had a profound impact on the next several decades of VPD research. And so I think this group was interested in coming up with an evidence base or coming up with a new definition, I should say. But people were just kind of suggesting ideas like, hey, let's define it this way. Let's define it that way. Why don't we throw in this and take out that and so on and so forth. And I thought to myself, you know, we really could answer a lot of these questions with objective data, we have big data sets now that they didn't have at their disposal in previous iterations. And so why not, instead of just sitting in a room and coming up with a definition, why not ask a question of whether you know, these things are going to be predictive of long term outcomes are valid or what not? So we went to that meeting, and then came back to chop, and every year, there's a cycle where you can submit an abstract idea to the NRM to use the determined data. And you submit a proposal sort of like you would have many granted, they will review it and say, Yes, you can do it or not. And so we thought, hey, this would be a nice way to use the NRN data until you look through the data that they have. And you say, hey, yeah, we could develop these kinds of definitions, and, and so on, and so forth. So we, we did that. But it's very serendipitous, that the definition that came out as being the best performing one also is one that's sort of convenient, and it's simple and whatnot. I mean, we could have come up with something that was entirely convoluted and difficult to apply. And we probably would still be trying to get the paper published. So I think it's a bit of luck, right? Like, it's it was, it was some insight that we said, hey, this is an interesting question. We could do it. And so you know, we'll get some credit for that. But the actual results of the paper or the data or the data, we don't get any credit for that. That's just how it, how it shakes out. And so it was nice that it was a good finding, or one that is a popular finding. But I didn't have any control over that.

Daphna: 16:52

That's fine. Go ahead. Yeah, no, I'm wondering, obviously, your definition, your papers definition isn't the the only definition of BPD. And I wonder how you kind of navigate that knowing that there are depending where you are in the world, depending where you're at what institution using different definitions of BPD? And how you see that panning out in the future. And certainly I think agreeing on a definition is valuable, so that we can continue to do research in the area.

Unknown: 17:27

Yeah, I guess, I guess there's sort of a two fold answer that. The first is that there have been a couple of external studies that I had nothing to do with. And we're not using interim data that have appeared to validate that what we came up with is a is a good definition, and it may well perform better than some of the other options out there. So I think there's reason for people to use it if they want to. But more importantly, we just need to be transparent, and what we do and how we report our results. So if you don't like the definition that we propose, that's fine, you can use something else, you just need to try and be transparent and how you define respiratory morbidity and why you picked it. And then that's okay, that's still valid in people through the peer review process, or, you know, later on after it's published with boots on the street can pick whether they believe the results to be valid. And if they want to apply that study in their own clinical practice, I mean, my hope would be that someone comes up with an even better definition than what we have. And to me, that would mean that we have new diagnostic tests, or we have new abilities to predict long term outcomes and to identify Sick Kids versus not sick kids. And so this should not be the last definition of BPD, we should hopefully have another one. And one that's, you know, based on some sort of objective diagnostic testing would be would be ideal. So I think we still need to do to do a lot of work. I, the thing I would caution against is the criticism that comes out with, well, why don't we just have let's count all the days that you're on respiratory support, or let's bring back some sort of physiologic test or, or let's have everyone get a CAT scan, or something like that, when I think these things represent things that the community really either has rejected, or is not yet capable of doing right. We've proven ourselves to be somewhat lazy when it comes to data collection and application of, of challenging definitions. And so I think it's valid to ask the question, would we be better off using something that's more complicated, but then I think we need to balance whatever gains we get in prognostic accuracy with the difficulty in obtaining that data and the difficulty of applying it uniformly all the time. Because if it gets complicated and people cut corners, then you're not really comparing apples to Apples anymore. That's right.

Ben: 20:02

And so for the for the listeners who I would be surprised are not familiar with the definition, right? I mean, the definition of BPD that came out from this paper is that you would grade BPD as either having no BPD if you're on if you're on rumour with no type of support, and grade one if you're on two liters or less of nasal cannula, great to if you're on two plus leaders, more than two liters, sorry, and or CPAP, or and IMV. And grade three, if you have invasive mechanical ventilation. I think the elegance and the simplicity of the of the of the definition, as you mentioned, is the fact that the Fit requirement doesn't really come into play in that specific definition. And I'm curious to hear your thoughts on when this definition sort of comes out of the of the numbers. Do you see it as saying, holy crap, FIU two is not there? And does that give you some happiness? Because it's more simple? Or does that worry you saying we should really have a fight or two in there? How come? It's not in the definition? So I'm curious to hear how the how did you deal with, with this finding yourself?

Unknown: 21:03

Yeah, you know, I'm, I went into the project, and I still am agnostic about all of those aspects of it. I you know, whatever work best is what we sort of agreed at the beginning, this is the way we're going to approach the the question, and we'll take sort of whatever we get out on the back end, had FY two been really important, then that means that it matters and that we need to use it. And we need to spend the effort that's required to calculate it correctly, and to record it correctly. And it turns out in this particular exercise, that it didn't add additional prognostic help, after you've already accounted for a baby's mode of respiratory support, I don't think it's fair to say that a child who requires different amounts of FY two is less is not less or more sick than another one, I mean, a baby who's on a ventilator with 100%. Oxygen, in my mind is very likely to be sicker than a kid who's on a ventilator and 21% Oxygen. But when it comes down to it, it's does that distinction matter when you're looking at a two year outcome? That's a yes or no. And the answer's no, at least in our analysis, if the question is, will the child on 100% Oxygen have worse Pft? S at two years? Or will they do slightly less? Well, on a five minute walk test at 10 years of age? I mean, I don't know the answer to that, but they might. But at some point, we need to simplify our endpoints. And we do that often with they have the endpoint or they don't have the endpoint. And in this case, in that, with that approach, it didn't matter how much oxygen you were on at 36 weeks. And so it's not to say that FAO two is not important, it's just for simply predicting the presence or absence of serious respiratory morbidity. Once you've accounted for the mode of support fo two doesn't matter. So I think there's still room for people in there to feel strongly about the importance of oxygen. They just need to recognize how it may help them to predict outcomes with greater granularity, and then we need to understand whether or not that gain in prognostic accuracy is worth it, you know, is the juice worth the squeeze? Basically?

Ben: 23:20

That's that leads me to my second question. And because I think this is a very modern approach that you have to data, right? I mean, you you trust the scientific process, and you say the data says what it says, However, having trained with Eduardo Banco ARIA at UNM, he always said this thing where he said, a patient requires an amount of FIU to but they only require it because you have them on it, have you really challenged them to see what they truly require? And have you tried to lower their fit, too. And so my follow up question to this discussion on that fire, too. Do you have an inkling whether the data set that you used showed that fit was not a factor? But could it potentially be because clinicians are not always sort of challenging their patients and that what the patients are on may not necessarily be what they require? And do you think that if we solve that issue, we might actually improve on the definition or make it more sensitive or more specific?

Unknown: 24:16

Yeah, it's a good question. I try to always be careful, at least when I'm writing papers to avoid stating that a child required this level of support or this level of oxygen, because really, they're they're being treated with it, they're receiving it, it was administered to them something like that, right? Because we don't know if they required it. And that can be true of a child on a ventilator, right? Maybe they could could be on CPAP. Maybe they would be better on CPAP. And so the same limitation could apply to modes of support and not not just FYI. Oh, too. So I think it's a very, very valid point. Certainly, the community has attempted to address that with the oxygen reduction challenge. The limitation there is that it's it's work right and this really goes back to my comment them out as being sort of lazy, sometimes we, in our data set, it was only applied about 50% of the time. And so it made it very challenging to use that information in a valid way, because there's biases in terms of the kids who undergo an oxygen reduction test and those who don't. There's a recent project. That's also from the NRM, that's essentially asked the question like, does the use of the oxygen reduction test improve prediction of long term outcomes? And so you have kids who are classified one way with the oxygen reduction test, and had you not done it, they would have been classified the other way. And then in that analysis, it really didn't. So again, it's sort of saying in the kid who's right along the line of maybe they have it maybe they don't does accurately distinguishing that make a difference for two year outcomes down the road, when you've said it's a yes, no thing and demand analysis. It really didn't. It certainly, I think, important that we come up with better ways to identify how much a child needs not only from an epidemiological standpoint, but also from a clinical perspective and the work that Eduardo's doing and others around oxygen titration, and automated oxygen level adjustments so that we're not constantly leaving these kids on too much oxygen or too little, it is sort of, you know, the next big important step and improving that, that that treatment approach. And so I think that it's an essential factor, or it will become a very important factor in how we treat these children. But in terms of differentiating need from requirement, and in a, in a research, sort of epidemiology type of study, I think it becomes very challenging, because we don't do the tests that are needed very often. And the subtle distinctions between those babies sometimes gets lost in in the numbers because they don't have a sufficiently important prognostic value.

Daphna: 27:04

I've always felt that the definition is maybe a little bit cleaner in terms of of patient care. When you talk about, you know, what are we willing to do and not not willing to do? You know, I think, if you don't rely so much on the oxygen, or you're willing to work with titrating, the oxygen, I feel like we're able to move babies forward, put them try trial them on lower support, certainly, as institutions are being compared to one another, based on some of these outcomes, as opposed to maybe keeping kids on much higher supports to keep an oxygen low to meet one of the other diagnostic criteria. I don't know what you think about that.

Unknown: 27:52

Yeah, I, you know, I have never quite certain when a child reaches a specific, you know, age threshold when I'm caring for them. So, I don't know that I come up to a kid who is at 35 weeks and change and I say, oh, man, we're getting really close to 36 weeks postmenstrual age, I don't want this kid to have BPD. So today, I'm gonna put them on less oxygen, or I'm gonna, you know, I'm gonna quickly put them on to low flow nasal cannula. I think that really that's the I mean, not to get too dramatic or but that goes against, like what we should be doing to care for these kids. Like, we should say, in my best judgment, this child needs X amount of support the positive pressure, or oxygen or blood pressure medications, or caffeine or whatever, right? Like this is what the evidence shows. But you're, you're not going to make that kid have a better long term outcome. If you turn down their oxygen on the day that they turn 36 weeks, those Minister leads, you might be getting your numbers to look better in a database or something like that. But you know, that's a false, a false approach to the care of these kids. But coming up with objective means to determine the appropriate level of support in children I think is important. I do think that sometimes though, we may we, we may err in the wrong direction. Like we often assume that that less is better. Right? So being on a ventilator bed being on CPAP. Good being on less oxygen, good being on more oxygen than in my clinical practice. I see sort of the opposite of that. I see the kid who's been on non invasive support for a long period of time who's been on high flow nasal cannula in their echo is getting worse and their brain metric peptide is rising and their co2 is have been 80 to 100 for probably too long, and they're not growing and they're not able to do physical therapy. And then they get transferred to our center. And the clinicians have been doing what they believe to be best, right? Because they believe that ventilation is bad for you that being on a ventilator is bad for you and being on a ventilator non invasive ventilation is good. But then that child has, you know, horrible pulmonary hypertension and they're not growing And they may well have been better off on an invasive ventilation. And so, understanding what is the appropriate support for a baby in either direction, right? There are too much or they're on too little, I think are are important areas of research that need to get sorted out. We have data on a prophylactic approach to using CPAP. You have a child who's spontaneously breathing in the delivery room, are you better off intubating that kid as routine practice? Or are you better off trying them on? nasal CPAP? The data suggests that a routine use of nasal CPAP is probably best. But after that, we don't really know from a trial perspective anyway, what are the best respiratory support practices as it comes to non invasive versus invasive ventilation? And when should kids be changed from from one to the other? So I think that we're applying data, good data, but we might be applying it to kids where it really wasn't generated in that scenario.

Daphna: 30:55

Yeah, we totally subscribe to that philosophy, then I have seen it in practice where, where, you know, people are trying to make the numbers work. And it's May it's maybe not the right thing. For the baby. We just talked about it on rounds last night, I said, if this baby's not tolerating his hands on, he goes back, there's no questions asked, even if he's oxygens not up, even if he's not to kick in again, if his if his status is changing in his tolerance to your care for him, then that's, that's what we have to have to do. And I think that's such a changing dynamic in medicine. I wonder if you had a crystal ball, and you could anticipate how you think the definition will will change maybe in the next decade? What do you think that might look like?

Unknown: 31:48

Yeah, I know, I. So I guess, in 10 years, I wouldn't be surprised if we're still using about the same thing. But my hope is that during that period of time that we acquire, and I think this is a reasonable hope that we acquire some objective imaging techniques that can be applied to infants with the most severe forms of the disease to help differentiate the sort of different phenotypes of the different subgroups in them. And so one example and I think one of the most likely ones that we might get is, is like objective computer automated reading of CAT scans to provide measures of, in homogeneity, to provide measures of air trapping, and then electrolysis. And combining that with things like Echo and maybe bronchoscopy and other tools that would help us to design treatments or study treatments in kids with the more severe forms of the disease, I think that we will likely be locked into a treatment based definition of BPD. For the kids with mild to no disease for the foreseeable future, because I don't think there's any point in taking a kid who's breathing and remember who was on two liters nasal cannula and getting a CAT scan in them putting in an endotracheal tube to look at their airway, it's probably reasonable to get an echo, at least my kid was still on some support. And that's a non invasive technique. But most of those kids are going to get better and will go home, it doesn't mean that they don't need therapies to help them improve their lung health or to avoid early onset lung disease when they're adults. But invasive or aggressive diagnostics, and I think should be limited to children with more severe disease. And I think there could be the possibility of very important gains and acquiring that information and then having therapies that can be tailored to those kids. So my hope is that we develop that technology over the next decade, but I suspect that in terms of classifying all babies, we'll still be left with a treatment based approach for the time being anyway.

Ben: 34:08

I was wondering, since we're talking about about next steps, I mean, do you I think you're very involved in the discussion about whether we should think about BPD and then neonatology and, and medicine in general as a Bayesian versus frequentist sort of field. Right. And, and the idea in many of your papers, you you'd like to emphasize the baseline risks and the importance of understanding baseline risks in the patient population that you're treating. I am wondering if you could explain to the audience a little bit what does that mean to have a Bayesian approach to research in in BPD? And how does that gonna affect how we can potentially even prevent BPD down the road? What does that practically looks like? Because I think this is something that is not being taught in med school, and that we and this is however, the way we all practice. We will practice Bayesian medicine without even knowing it, and all the papers that are coming out are frequently this type of paper. I'm wondering if you could talk a little bit more to that?

Unknown: 35:08

Sure, I'll do my best. So I'll get the caveat to all this is that when it comes to statistics, I'm only skilled in the frequentist methods, and not all that skilled at that. So I can't provide you an explanation of how to do any Bayesian math. But the concept of it, I think, is very appealing, because it, it fits with the way we think about our approaches to treatment, right. And we basically say, this kid looks healthy, this kid looks kinda sick, this kid looks really sick. And so if they're healthy, all the things that I'm doing right now are working, and I can either keep up what I'm doing, or I can do less. And if this kid looks kind of sick, then I should, you know, think about some diagnostics to determine whether or not I should pick a therapy or not. And if this kid looks really sick to me, then I need to add new things, right, they need new support, they need this or that to help make them better. And ideally, that's how we would identify children for our future studies. And it is how we do to a certain extent, right, we say, you know, we want to study nasal CPAP, in the delivery room versus invasive ventilation. And so we make entry criteria, we say you need to be 24 weeks to 28 weeks, we wouldn't enroll a kid in that study who has a gestational age of 35 weeks, because their baseline risk for BPD is is essentially zero, we're very close to that. And so all of our studies, to a certain extent, employ the concept of who's at risk for the disease, or who's most at risk for the disease, and then we'll enroll them into that. But we have, in general, take a very broad approach and just said, Okay, you're extremely preterm infant, or you're an extremely preterm infant who's on a ventilator at this age. And so you're, you're at high risk of developing a poor outcome. And so I'm going to enroll you in into this study. But I think as it pertains to BPD, the more we've learned about the disease is that it has a very heterogeneous pathophysiology, both in its development, and then in its presentation or clinical manifestations. What when the children have sort of developed BPD are at risk of developing it. And so understanding the baseline risk, or how much risk they have of developing the illness, and why they have that risk, is it because of their lung abnormalities or pulmonary vascular abnormalities or so on and so forth. And using that information to help direct our therapies, and designing trials that include more selective groups of infants to study interventions that we think will most likely be beneficial in that sub cohort? Think many times we get therapies that have point estimates that favor use, but the confidence interval is too wide. And it might be that they were underpowered, but at least for our field, they feel like big studies, they have several 100 kids, and then if that, if not more. And so I often wondered, did that intervention fail, because it really doesn't work. But when the confidence interval is just a little bit above one, you sort of wonder, well, what happens if we didn't rolled less of this type of kid and more of this type of child would we have said, hey, it really works, but only in this subset, because they have a baseline risk, based on their physiology, or they're just at the most risk or the least risk. And so those are tough studies to do, because babies are hard to enroll. And so we need more centers and longer enrollment windows, which is challenging from a funding perspective. But we select therapies every day, based on the baseline risk of a child inherently in our own minds, right? We say Oh, this kids, you know, they're not doing so hot, I want to start steroids or this kid is, looks like they're having more athletes. So I'm gonna give caffeine right. So we know that that child has a higher risk of developing a poor outcome. So we treat them accordingly. So if we could come up with ways to better design our studies around that, I think we would find therapies to be more effective, but they're just more effective in some groups of infants, and not among all extremely preterm infants, or whatever.

Ben: 39:35

That's, that's great. That's great. Because it's exactly right. I mean, at the end of the day, it's it's exactly the thought process is there in every step of care whether it is at the bedside and research and yet, it's something that somehow is not being addressed more frequently. Definitely want to do you want to do something or can I go ahead?

Daphna: 39:53

Well, I wanted to make sure you you got in all of your Bayesian questions. So thank you. So my question is really about kind of the ongoing care of these, like you said, you see the sickest BPD babies, and they're not for the faint of heart and not on neonatologist, like managing these very sick BPD babies. And so I'm wondering if you can tell us, you know, kind of your approach to the very sick baby, especially because you get so many transfers in so you get babies, you don't know very well, just that they're sick, and how you kind of move forward? And in deciding, you know, respiratory support interventional therapies, because, you know, I you do, I think not a one size fits all approach. And so, you know, what tips can you give people who are not quite comfortable managing those very sick babies other than sending them to a center like yours?

Unknown: 40:59

Yeah, sure. So I think that I think that people tend to, if they have some trepidation about caring for those kids, that it may relate to the fact that they don't see very many of them. And related to that, is that the focus, I think of caring for these kids is in some respects, sort of a 180, from what it is, and caring for the newborn, extremely preacher. In that population, might be one of the few circumstances where sort of the Bayesian approach is less employed. Because we say, All right, the goal should always be to put the kid on to non invasive support, we want to give her fact and if we're going to intubate and then we're going to start caffeine. And we're going to limit their fluid intake, we're going to try to increase feeds fairly rapidly. And so there is a bit of a playbook and how to care for the newborn child in the first week of life. And then after that, hopefully, the therapies that you've initiated have worked. And then we can begin to gradually wean down. The difference, I think, between that mode of thinking and the mode of thinking for a kid with established BPD, who is sick on high, either non invasive support or on a ventilator, is that they fall into many different buckets in terms of pathophysiology. And so an approach to weaning aggressively might work in some child, but it might be the exact opposite of what you should do in another. The disease is chronic in nature. And so getting a blood gas every day may lead you just to go up and down and up and down and up and down on support and not really see demonstrable gains, their primary problem might be airway related, instead of lung parenchyma. Their primary problem might be lymphatic or infectious and so employing the tools that we have available to us and understanding their limitations to try to come up with individualized treatment plans in these kids is a is a learned skill in two regards, one, think you have to take care of a bunch of them to develop those skills, but to it's a skill that we have yet not yet perfected by any means. And I think in many respects, we have not yet even gotten to the point we can say we're good at it, I think we're getting close to being good at it. But we just don't know when to do the test, who to do them in what tests are good, how valid they are. And so all of that uncertainty leads to a lot of trial and error. I think that's uncomfortable for clinicians. It's uncomfortable for me, certainly, it's uncomfortable for parents. But I think that it's just a collective gain of experience. And it, it's another, maybe not quite Bayesian, but it goes to this, you know, I don't quite know why this works in these kids, but it seems like it does. And so we're going to use it and try to tweak it a little bit more and, and tailor it to this kid. And you know, I don't think the oscillator is a good choice in this kid. I don't quite know why, but I've seen it that work, or it has worked here there. And so it can be difficult to you know, if we get a console in from an outside hospital, it can be difficult to say, alright, I've looked at the X ray, I've looked at your lab values, and I've looked at your P co2 levels, and so on and so forth. And this is how you should change the ventilator. Because you sort of have to have that kid in front of you and look at their breathing and put your hands on the ventilator and turn things up and turn things down and see how their chest rise changes and listen and whatnot. And, and those are things that I I can't describe clearly in words, but you just sort of need to do them and I learned them by watching more senior folks at my institution do them and then by having my own trials where I had certainly errors and kids didn't do well as well unfortunately and other times where I've tried something that people hadn't shown me before I said hey, well this this really works in this baby Why Why do I think that isn't so I think that level of uncertainty and accepting that a kid might get a little bit worse while you figure something out to get them better is necessary to help each child but it's an uncomfortable place to be as a clinician where you come from an environment where there is this sort of playbook as to how you take care of a kid, and even if they're getting worse, right, so a newborn baby, they're not doing well on a ventilator. So now we rescued the oscillator. And we we started meeting airway pressure, that's two points above what it was on a conventional ventilator, and we target a P co2 of 50 to 65, or 50, to 70, whatever your institution is, you know, if I have a kid who the best co2 I get out of them is 90, but they're growing and like aren't that this best I'm going to do. So that's okay. And you have to get comfortable with that level of disease that we may not accept in a smaller child.

Ben: 45:47

I think the term that I've used for what you described earlier is tinkering, right? These babies require a lot of tinkering. You need to go at the bedside and just play and just go up and down. I think that's yeah, that's that's that's not a term that I came up with, obviously, I mean, but I wanted to ask you about managing a little bit some of your career. I mean, I think a lot of us young neonatologist look at you and career and sort of this video, Eric Ries that you've had in such a short amount of time. And I'm wondering, in terms of balancing both clinical research family life, what are your tips for for, for other young clinicians? And I was wondering on a practical note, is there like any tool or software or mobile app that you use that you recommend? And that has been a game changer for you that others could use?

Unknown: 46:35

Yeah, so there's a couple of questions, I'll try to answer as best I can. I've been been fortunate to work at an institution that has, from the get go been generous and protecting my time. So before I had grant funding, I only had to do 12 weeks of service, and there was internal money that could support me not doing for us full time is 20 weeks. And so I was a little above 50%. And that allowed me to get started on things. That research training that I had was also very important, because one of the things that limits young folks, I think, is an inability to do the math for their projects, right, they have access to a local database or a larger multicenter database or something, but then finding a statistician or even knowing a statistician that you can talk to and then say, can you do this, and can you not take a year and a half to do it? Is is challenging, right? It's intimidating. So you're almost you just say, oh, man, I'm not gonna do that. Or you'll do something simple that you can do in Excel or whatever. And that's all fine. But in order to get higher impact papers, and things that I think get closer to answering the questions that we want, you need to do more robust math. And so I was fortunate that I was head training in order to do some of that on my own. And I've, I'm a person who enjoys doing the math side of things, I think, a bit more than some of my other colleagues. And so I'm more likely to sort of get into the weeds and play with the numbers. So I think that's been helpful for me. I find it to be useful to sort of think about, like moving projects, sort of off my plate and onto somebody else's. Most things, you're not brilliant, you're not doing on your own. And so you don't need to think of alright, I'm just starting this, and I have to get it published, right. The first thing is like I need to get like a table or figure I can show to somebody and have them look at it, give me some feedback. And then you need to get some paper written down, you know, you need to get some text on the paper. But after that, you need to send it to somebody else to read it. And after you do that, it's no longer your problem until they give it back to you. And then you can do something else. And once they give it back to you, then it's your problem again, and then you send it off to the journal to review and then it's not your problem anymore. And then you go about doing other things. And then you know, they get it back to you. And so just thinking about not how am I going to get this published per se, but how am I going to move it from its current state to the very next state? has, I think been helpful for me to just compartmentalize the what I need to get done today. I think in terms of you know, I'm not I'm not particularly tech savvy, so I don't use I still print out most journal articles. So if I'm going to write a paper, I printed out a big stack of journal articles and I highlight and I put little comments in the corners and whatnot. And then I have a file folder full of just you know, lots of papers that I'll probably never go back to so it's it's not good for the trees but I I do things a little bit more old school, I guess in that regard than some younger folk do. But I am a huge fan of using Adobe Illustrator. If you'd like to tinker with images or graphs and things. I really like you Using that, because it can make the graphs look a little bit nicer, you can format them the way you want. And so if you are out there doing your own research, and you want to make images that I think pop a little more than just simple bar graphs and things like that Illustrator is a fairly easy tool to use. And there's sort of web videos that you can watch. And so that's probably the one techy thing that I've adopted, that I find to be helpful. And I think that if you have a nice color figure, or a nice, you know, graph, or something that really tells the story of your paper, it's easy to convey, people want to put it in a PowerPoint, and like, you get just a slightly higher leg up in terms of, you know, journal selection and whatnot, because we're, we'd like to see things that look good. And so I think the more you can sell your research with, easy to see figures, it's very helpful. And I've found Illustrator to be a very nice tool for that.

Ben: 50:55

I think that's a great advice. I could not agree with you more. I mean, every time. I mean, we do journal club every other week, and like if if we could have a nice figure that tells the story of the paper, it is worth 1000 words. And so I agree with you. And I think this is something people should do more more often. We're coming close to the end of our hour. And again, I wish it was longer. But I'm gonna let Daphna finish off on on some of her questions before I ask my last few ones.

Daphna: 51:21

Well, I wanted to touch on one more facet actually BPD care, you said something early on about what attracted you to pediatrics. And that was that you actually liked talking to parents. And I wonder for some of my colleagues, one of the things they don't like about BPD care is that those are frustrated parents, right? Those are parents who have been in for the long haul, they've seen their kids get sick over and over and over again. And they're, they're tired of the tinkering, right? Nobody, they feel like nobody knows what they're doing. We're just trying different things on their babies. So I wonder kind of what's your approach to the parent who you know, is, is exhausted and their baby has BPD. And you know, that it's going to be more months until you know, they get to take their baby home?

Unknown: 52:17

Yeah, I think I think one of the most important things that I've found that works well, for me is just honesty. And telling families when you don't know, telling them, your level of confidence that your choices will work, or that they will work and why you are making them. I find that involving them in some of those decisions. It depends on the family, of course, but involving them in those decisions can be very powerful. If you say, you know, we can keep doing what we're doing now. And this is the level of success we're having. And you're seeing your baby having these these ad spells, or these things that are unfavorable. Alternatively, we can try a different ventilator. And it may help it may not help. I've seen it work in these cases and not work in these cases. And if you're willing to let us try that, I'm happy to do so. But if your preference is that we stick with our current state of affairs, then then that's what we'll do. And I'm okay with either decision. And so they're absorbing a bit of that decision, as well. The other thing, I think that we have a tendency in medicine in general, to wanna give good news. And I think that in the care of children with BPD, we often have to give not a lot, it's not bad news. But it's, it's like less exciting news, right? So how long is the child going to be in the hospital or if they come to our center at 40 weeks postmenstrual age will probably be there for another five or six months. And they're, they're often shocked to hear that number. And so I'll say, Oh, well, this is, this is how I came up with that, right, your your baby's here, that's going to take me a month to get them to hear and then it's going to take me a month to get there. And then it's a week after that each week, I'm gonna do XY and Z. And that's the best case scenario. And already, you can see how I've come up with four to five months out of this. And then they say, I sort of I, I kind of get that. I think it's okay to acknowledge that we don't know what we're doing. We have experience in these kids. But we're, you know, we're trying the best that we can and many children follow the rulebook, right? And we've talked about a couple of times that, you know, we, the first week we do this, and then we do the weaning, and that's what happens with most kids. That's why we do it and it works. But there are kids who don't follow that rule book and your baby happens to be one of them. And I can't change that. But what I can do is work with you and your child to help have the best outcome for that baby as I can. I think that level of honesty and sort of working through why we're doing what we're doing and how it might be different But from what they've encountered at another hospital, it generally works, families are still going to be frustrated. And that's okay, I'm happy to be there sort of verbal punching bag when they're having a bad day, I've got thick skin, and it doesn't really bother me. So if they want to say you guys are dumb, and you don't know what you're doing, and so on and so forth, just like lay into me for like, 10 minutes, that's okay. And then, you know, after that passes, we can talk about what we're going to try to do, I can tell them that the vast majority of babies survived discharged from our hospital. And that's my expectation for your baby. Everything after that is far more uncertain. So we're going to have to work through those details as they come to us.

Ben: 55:36

I think it's nice to hear that type of humility from Dr. Jensen who, who also says that it's okay to say we don't know what we're doing. Because that's how most of us feel. I mean, I guess, following up on that, I mean, I like to ask that question to people who have encountered a lot of success, because I think it humanizes the individual. And so I'm going to ask you this question to you as well. What's in your career? So far? What's been your most memorable rejection?

Unknown: 56:02

Most memorable rejection? Well, I got. So I submitted my first K grant. And it was I have one, not even that one. But I am I frequently submit grants that are not discussed. So I think for the people who know, the grant submission process, right, you you spend all this time writing a grant, and you think you've written something really good. And especially when you're young, right? It's not like you've submitted a whole bunch of grants. And so you're just sort of playing the roulette wheel of research, you're like, This is my grant, this is my project, and like, I'm staking my career on this, then you submit it in and then the people who review it are like, this is so much garbage, that we're not even willing to talk about it at a meeting. And yeah, wow, that's a huge gut punch. And so I submitted my first grant. Okay, Matt, disgust I submitted. And the reviews I got were like, we don't think you can do any of this stuff. This is nonsense. And my mentors were like, This is a great grade, and like, this definitely gets funded. And so you know, that's, that's heartbreaking and embarrassing all at once. And then I started like, Alright, I'm gonna submit a completely different one. So I submit a brand new que, submission, and also not discussed, like, wow, these people have zero interest in talking about my work. And so eventually, on the third try, I got one funded, and then sort of worked on that for a little bit. And then I submitted my first sort of independent grant, I submitted an AR 34, not discussed, like, so it's just every time I submit a grant, it's like, they're probably not going to talk about it. But thankfully, I resubmitted that, that independent one, and it got funded. And I've been able to get a few others funded as well. And so you know, it's just, it's a battle of attrition, right, and you hear this from the NIH folks all the time. And I think that's true in everything that we do in academics, it's like the people who are willing to persist, are going to have success, right, you will get lucky from time to time. And I think I got lucky with the BPD definition project, for sure. But that's not enough to get a grant out of it, right, in order to get funded, you just need to keep hammering at the nail. And I've had a lot of failures. And I know people who have been doing this a little longer than I have, have had many, many more failures, and they just persist. And so you just let it bounce off you you learn from the critiques you take the good and the bad, and then you keep moving on. And so I've I've tried to embody that in my work, it's not easy, because right we take, we take a lot of pride in what we do. But you know, you you get the rejections back you do whatever it is that you do you have a have a drink or you know, play with your kids or whatever. And then you know, you let that pass and then you move on.

Daphna: 58:44

Now, as you say, having thick skin and in patient care and academia then

Unknown: 58:51

probably all in medicine.

Ben: 58:55

Thank you for for sharing that with us. And I don't want to end on this note, obviously. So my last question to you for today is is if it had not been BPD, what are the areas of neonatology really sparked your interest? And you don't have to go too much into depth, but just like what are the other aspects that you're like, oh, man, if I wasn't doing BPD research, that's kind of cool.

Unknown: 59:13

I like I don't know that it's a particular disease. But I like sort of disproving commonly held notions that are just things that we believe to be true. So my pet project is the carseat study. It's something that we've been doing for 30 plus years, and there's really very little data to support its use, it doesn't mean that we shouldn't be doing it. It's just that we've sort of taken for granted that what we're doing is correct when it may not be so I think if I were to have another passion in neonatology, it would be to think about practices and diagnostic tests that we have adopted without demonstrating that they're useful and studying whether or not they really are and it really sort of thrilling out the stuff that we do that is wasteful, because it's expensive. And it can be frustrating. And it could be causing harm. And we're not intending it to because we're coming at this with the best intentions. It's just that we don't know that what we're doing is wrong. And so think, studying that those types of questions, how can we improve care by by doing less of what's wasteful? And what is unvalidated would be an interesting area. Thank you.

Daphna: 1:00:28

Very cool.

Ben: 1:00:30

Definitely anything else?

Daphna: 1:00:32

Now, I just want to thank you for for spending this time with us. We know you've got a busy schedule, and I think that our listeners, particularly our trainees, I think we'll get a lot from this episode. So I really appreciate your time.

Unknown: 1:00:47

Thank you for having me.

Ben: 1:00:49

Yeah. Thank you so much for being on the show. I think this is going to be a popular one. And thank you for your humility and for sharing all these stories with us. Yeah, thank you so much.

Unknown: 1:00:59

Thank you guys. Have a good day.

Ben: 1:01:02

Thank you for listening to this week's episode of the incubator. If you liked this episode, please leave us a review on Apple podcast or the Apple podcast website. You can find other episodes of the show on Apple podcasts, Spotify, Google podcasts, or the podcast app of your choice. We would love to hear from you. So feel free to send us questions, comments or suggestions to our email address NICU podcast@gmail.com. You can also message the show on Instagram or Twitter at NICU podcast. Personally, I am on Twitter at Dr. Nikki spelled Dr. NICU. And Daphna is at Dr. Duffner MD. Thanks again for listening and see you next time. This podcast is intended to be purely for entertainment and informational purposes and should not be construed as medical advice. If you have any medical concerns, please see your primary care practitioner. Thank you