#385 - Dr. Andrew Beverstock - Exploring Neonatal Nutrition: The Role of Urinary Sodium

Hello friends 👋

In this episode, Dr. Andrew Beverstock discusses his research on urinary sodium and its relationship with growth in preterm neonates. He shares insights into the importance of sodium for neonatal growth, the methodology of his study, and the unexpected results that challenge existing literature. The conversation also touches on his diverse medical training, mentorship experiences, and his involvement in medical education and point-of-care ultrasound (POCUS). Dr. Beverstock emphasizes the significance of careful population selection in research and outlines his future research directions.

----

Short Bio: Dr. Beverstock trained at the University of Edinburgh in Scotland, where he completed an intercalated Bachelor of Medical Sciences in Developmental and Cell Biology. During this time, he worked in Dr. Sally Lowell’s laboratory at the Centre for Regenerative Medicine, focusing on the differentiation competence of pluripotent stem cells. He then completed a two-year categorical internship within the UK Foundation Programme at the University Hospitals Bristol and Weston NHS Foundation Trust, followed by a one-year pediatric teaching fellowship at the Bristol Royal Hospital for Children, during which he earned a Postgraduate Certificate in Medical Education.

Dr. Beverstock subsequently moved to the United States to pursue residency training, which he completed at the NYU Grossman Long Island School of Medicine in Mineola, New York, before relocating to Houston for fellowship. He currently serves as a junior author for the NeoReviews NeoQuest segment. He is the recipient of the Evangelina Whitlock Memorial Grant, which supports his clinical research project examining urine sodium and urine sodium-to-creatinine ratios in infants with and without faltering growth. This nested matched case-control study investigates whether urine sodium values differ between infants with faltering growth and whether urine sodium correlates with growth metrics.

----

The transcript of today's episode can be found below 👇

Srirupa (00:00.847) Hello everyone, welcome to another fantastic episode of Rupa's Fellows Friday. I have with us Dr. Andrew Beverstock from Texas Children's Hospital. He's currently a second year neonatology perinatology fellow and a very dear friend of mine. I'm very excited to hear all about his investigations on sort of breaking the taboo field of urine sodium. I'm very excited to hear a lot about what he's done so far, what his initial results have shown and what his plans are and his next steps are for his project. He's also someone who's very interested in medical education and point of care ultrasound in the NICU. And I would love to hear his experience in establishing goals for POCUS (Point-of-Care Ultrasound) program at Texas Children's as well. Welcome to my show, Andrew, how have you been?

Andrew Beverstock (00:49.228) I've been great. Thank you so much for having me today, Rupa.

Srirupa (00:51.909) That's fantastic, and I'd love to hear about your project. I know when I graduated out of Texas Children's, Andrew was beginning to create his specific aims and begin the initial putting out of the seeds here and there about his project. And I'm so excited to hear what that turned out to be and what the results are and what the initial results are. So share with us a little bit about your project and what you found so far.

Andrew Beverstock (01:15.692) Of course. So I'll talk a little bit first about the background to my project. I know that not everybody may be familiar with the use of urinary sodiums. And so I'm going to talk a little bit about that. So I've been lucky enough in fellowship to complete a project looking at the use of urinary sodium and urinary sodium to creatinine ratio and the relationship with growth in preterm neonates. And I had not come across this in residency and then came across this in fellowship as a concept and decided to study it.

We know that sodium is incredibly important for postnatal growth in babies. It has a lot of roles in growth. On its most simple level, babies need sodium to make DNA and make new cells. And we know that sodium supplementation helps babies grow better. That's been pretty extensively studied back in the 1990s and then onwards from there. Giving babies additional sodium tends to make them grow better than if you leave them on unfortified or non-supplemented milk.

We also know that sodium has a complex role in other physiologic processes that can help babies grow. Babies need sodium for glucose uptake in the small intestine. Yes, SGLT-1 (Sodium-Glucose Linked Transporter 1) requires sodium to function. It also has increasingly been found to have a role in the development of the small intestinal microbiome, which we know is extremely important for growth. We know that babies who have a higher sodium intake have more lactobacilli. Lactobacilli obviously have probiotic effects and that can help babies to grow more effectively.

And we also know that it has a role in bile acid recycling through the apical sodium dependent bile acid transporter, which alters the ratio of conjugated to unconjugated bile acid uptake. And so we know that sodium has all this multitude of roles that allow babies to grow, even though we know that also that too much sodium can be harmful for babies. And so there is an important balance to be struck there. We know that giving too much sodium can make babies hypernatremic.

And there has also been some evidence that it can cause delayed gastric emptying. And there may be an association with necrotizing enterocolitis, particularly through the mechanism of raising feed osmolarity. So it's something that we need to be cautious with. We know that there are benefits to it, but there has been a lot of work in recent years trying to determine the optimum candidates to give sodium to. And we know that preterm babies are particularly at risk of sodium deficiency. They can often have a low dietary intake of sodium and they also have very high renal losses of sodium since they cannot adequately absorb sodium in the kidney until they're around 33 to 34 weeks gestation, which is when that tubuloglomerular feedback mechanism matures. And so these babies and our preterm babies, especially our extremely low birth weight infants are at high risk of becoming deficient from losses. Also, we know that they are at risk of having insufficient intake. Often our babies are initially on unfortified milk and we know that unfortified milk does not meet the sodium requirements for preterm babies and particularly those babies who are fed donor breast milk which comes from moms who are at the end of their pumping journey who have mature term milk and that milk is extremely low in sodium compared to colostrum which can have 20 or 30 times as much sodium as that donor milk.

Some babies don't tolerate fortifiers as well. At Texas Children's, we use human milk-derived fortifiers, which are obviously a great product, but those also tend to be low in sodium. And so we know that babies are at high risk of sodium deficiency, and that may contribute to growth failure in these babies. And so there's been a lot of research over the last five years looking at how we best identify babies who would benefit from sodium supplementation. We obviously want to make sure we capture babies who are growing poorly because they are sodium deficient without including babies who are sodium replete and putting them at risk of those harms of increased risk of NEC (Necrotizing Enterocolitis) and that delayed gastric emptying and feeding intolerance. So I think all this work initially came from looking at babies with intestinal failure. So we know that babies with short gut have high sodium losses, particularly if they have ostomies. And there was a paper looking at the relationship between growth and urinary sodium in babies with intestinal failure.

And what that paper found was that having a high urinary sodium was correlated with adequate growth in that population. And so the thinking is that if babies are sodium deficient, they absorb, resorb more sodium from the urine to try and maintain their total body stores. And that has been very robustly studied and has shown to have a strong correlation in these intestinal failure babies. And so when we were looking at, when I started fellowship, we were using this in preterm infants as well. Now, Texas Children's, we have a large number of babies with intestinal failure. We have a neonatologist run intestinal rehabilitation program run by Dr. Premkumar and Dr. Hair. And we were using it in that population with great success to look at why these babies were growing slowly. But it was also starting to creep into our preterm population where we were looking at urinary sodiums in babies who had slow growth. And using that to determine whether we did or did not give them urine sodium. And there has been a number of publications on urinary sodium algorithms using theoretical calculated values in the preterm population to determine who gets sodium supplementation. I think the most notable of those comes from the University of Iowa, who have published pretty extensively on this topic and have shown that their urinary sodium algorithm, where they give more babies sodium, leads to better growth in that population.

But we were a little, we wanted to make sure that that wasn't just therapeutic drift and that we were correctly identifying the best babies there. And so we, there were reasons I think to be a little skeptical of urinary sodium in this population. I think there's something of a contradiction where we know that these babies are partially sodium deficient because they have high urinary sodium losses because they cannot absorb sodium. And so how can you say that they're sodium deficient because they are not reabsorbing, but also that urinary sodium is a reliable marker of their total body sodium status. And therefore we decided to do a prospective study to try and evaluate whether that same relationship exists between growth and urinary sodium in the preterm infant population as it does in the intestinal failure population. The other thing that's been studied in the intestinal failure population that has never been looked at is this urinary sodium to creatinine ratio.

And there was one small paper from a few years ago that found that having a very high urinary sodium to creatinine ratio actually correlated very well with growth as well. The problem with urinary sodium and serum sodium is that it's very much affected by your volume status. And the theory is that if you introduce creatinine into the mix, you can better account for that volume status and perhaps have a more true reflection of your total body sodium status. And so that was the original basis for how we came up with this project idea and why we decided to study this, which I think is a fascinating topic and one that is well deserving of study. And so we designed a case control study. One of the things we're lucky enough as fellows at Texas Children's is to have access to some research funding. And we wanted to make sure, urinary sodium can be an expensive test to do. It's an ELISA (Enzyme-Linked Immunosorbent Assay). And we wanted to maximize the number of babies we could study and the efficiency of that study by ensuring that we use the right mechanism and the right study design. And so we decided to do a case control study to look at babies who were growing very well and babies who were growing poorly and see whether the urinary sodium and the urinary sodium to creatinine ratio differed between those two groups. So our study included 30 infants. We had 15 babies with faltering growth and 15 babies with adequate growth. And we compared the longitudinal urinary sodium in this group.

And I think one of the things that we were lucky to have was samples from different time points for these babies. We used banked urine samples from these babies. We had about eight samples to choose from in each group. And so we made this a sort of pragmatic study where we looked at the time points that people are looking at in real life to see whether that correlation was there. We ended up looking at the two week mark, the three week mark, the four week mark, and then the 34 week mark to give us a post-renal maturity time point to compare to.

And we analyzed the urinary sodium, we analyzed the urinary sodium to creatinine ratio, and we used mixed effects linear modeling to compare whether these were different between each group. So I want to emphasize first that our results are not yet peer reviewed. We are still working on the last of our data, but so far, based on the data that we found, we did not find a correlation between growth and urinary sodium or urinary sodium to creatinine ratio within this population, which is a little bit different to what the theoretical basis should be and what the data has shown in intestinal failure infants. And so while we were hoping to prove the utility of urinary sodium, I think this has raised some questions about whether it is such a reliable marker in that preterm population.

Srirupa (10:29.388) No, that's fantastic. And thank you for sharing with us about your thought process and the basis of why you chose this project. I think I always say this, that null results are important. Like especially contradicting results. Because when you see an abstract, a paper, a peer-reviewed publication, the automatic thought is, is this a single center application? Or can this be extrapolated to other centers? You sort of strengthened that thought in me by saying that okay, I have a contradicting result here. And it's a result that's not what you thought and it's not a result that reflects what's in the literature as well, which again makes you think are single center studies applicable to all the babies? Because you do have differences in the way you practice and the way babies behave and in the physiological transitional things that happen in babies as well. I think that part is very interesting. I see that, and I failed to mention this to start with, but you've had a very diverse experience beginning in your medical career. You did your medical school in Scotland, and then you moved to the United States after you did your teaching fellowship at Bristol Royal Hospital for Children. And my question to you is that with this diverse beginning in your medical career, did you see something similar to the use of urine sodium in your previous training periods? Like did you see this in the UK for example, in Scotland for example?

Andrew Beverstock (12:05.518) So that's a great question. So I had a slightly unusual path to neonatology where, as you said, I trained in the UK and in the UK we do a sort of categorical internship where you rotate around a bunch of different jobs once you graduate. And so you have to do a bit of adult surgery, a bit of adult medicine. And I actually ended up not doing any pediatrics within that internship. I did pediatrics very late in medical school. It was my last rotation. I had already locked in my jobs by the time I did a pediatric job.

So I actually did not even rank pediatric jobs. It was not even on my radar as a career I wanted to consider until I did it in med school, at which point I had locked into this two year kind of cardiology stroke, upper GI surgery type, make me an adult acute medicine doctor. And so I did not come across urinary sodium in the UK, primarily just because I didn't work in neonatology there. And so it was really not until I became a resident that I did any neonatology work at all. I think neonatology is a fairly niche subspecialty within pediatrics, particularly in the UK. We did probably a day of neonatology at medical school and then none since then. And so this was not really something I had come across. And even in residency, I did residency from 2020 until 2023 when this urinary sodium work was starting to get off the ground. And even then we did not use it in residency and I wasn't really familiar with the literature.

I think since, we started planning this project back in 2023. And since then there has been a number of very great publications that have come out looking at pre and post implementation of these algorithms. Iowa has published the results and they had really fantastic improvements in growth and also some slightly unexpected findings in their paper. They found fewer infections, they found less time on mechanical ventilation with this. And so it really highlights the complexity of the role of sodium in preterm physiology. They raised the possibility that maybe there's some role for sodium in immune function there. And so it wasn't really something I'd come across, but certainly it's a field that's ripe for study. And I think it's when you have unexpected findings like Iowa did, it really brings you back to that translational part of research where you find the clinical finding and rather than taking something from the bench to the bedside, you take it the other way. You go back and look at how you can bring it from the bedside, going back and looking at the mechanisms in animal studies and lab studies instead.

Srirupa (14:34.04) Yeah, no, absolutely. And thank you for sharing your journey because I think that that's so fascinating to kind of see, you've seen and dealt with different aspects of medical care here, if you will. And it's brought you to Texas Children's with a whole new idea to explore and investigate. So that's fantastic. Share with us, Andrew, about what your next steps are for your project. What would you like to imagine your project to bring out in the next few years?

Andrew Beverstock (15:00.046) Of course. So we are still finalizing the last of our data intake. We have run our analysis with the mixed effects linear modeling, thanks to our excellent statistician, Joe Hagen. And we are looking at just looking at a final few factors. And so one of the things we wanted to look at was comparing caloric intake and total sodium intake for these babies. We have standardized reference ranges for the sodium content of donor milk and we can use standardized ranges for breast milk to compare how sodium intake compares to urinary sodium output. One of the things we know about urinary sodium is it's regulated by gastric secretion and by endogenous dopamine generation. And so we know that your dietary sodium intake controls your urinary sodium. And so one of the things we want to look at is to see whether those babies who received a higher amount of sodium in their diet perhaps because they were being given mostly maternal milk rather than donor milk, whether that affects your urinary sodium output. And so that is one of the things we want to look at next. And then the other part of our project, which I won't discuss too much since we're still in the very early stages, is we're going to also look at the relationship with bile acids. And we talked earlier a little bit about the role of sodium in regulating the apical bile salt transporter in the small intestine. And we're going to be looking at bile acid profiles in our babies and to see whether they differ between babies with faltering growth and adequate growth and whether there is any relationship with sodium intake there as well. And so hopefully we'll be able to get all of that data analysis done in the next few months and then start to present our work at places. Because I think this is going to be work that generates a lot of discussion. There are people who feel very strongly that this is something we should be using. There are people that feel very strongly that this is not. And we're hopeful that being able to take this and present this at different national conferences will give us the opportunity to get that feedback, generate those discussions and hopefully move on to further work. With the long-term aim being that I would love to do a prospective study looking at following babies that we give sodium supplementation to to see whether that changes their urinary sodium output. And that's a difficult thing to get done in three years of fellowship where you have limited time and unlimited budget. But I would love to continue that work as I move forward in my career, because I think this is such an important topic. There are so many babies who have this idiopathic postnatal faltering growth, and it would be great to try and identify those who are in an indeterminate way to make them grow better and find the babies that are going to benefit from sodium supplementation.

Srirupa (17:34.969) Yeah, absolutely. And I think that's a fantastic goal to have in the next few years. And I think that you bring up an important point that these are the studies that generate a lot of discussion because, like you said, there are people that are very much in favor of getting the urine sodium values and clearly have shown some benefit with that as well. And there are people who don't feel very strongly about it. So these are very important studies. And anything that generates a discussion is always up for multiple, multiple investigations in the future, for sure.

Fantastic, Andrew. Tell me a little bit about, you have Dr. Hair as your mentor, who is one of the big pioneers of neonatal nutrition. So tell me a little bit about how you chose your mentor, how you built that relationship with Dr. Hair. I'm a big fan of Dr. Hair myself, and I would love to hear your thoughts on how that came about happening.

Andrew Beverstock (18:26.552) So I was very fortunate to come across Dr. Hair's work prior to starting the fellowship. So at the end of residency, I was on paternity leave. I had just had a baby. And one of the things I spent my paternity leave doing was actually listening to some Incubator episodes. And I came across Dr. Hair's work. And it was her and Dr. Misty Good talking about their neonatal nutrition work at that point. And I was familiar with her work through that.

And then Baylor has a very good system where we have a huge faculty here at Texas Children's. We have, I think, more than 150 attendings across all of our sites and more than 50 on our main campus site. And so when I first started fellowship, we were sent a list of faculty and their research interests and whether or not they were taking mentees. And so I came across Dr. Hair's work that way and set up a meeting with her during our summer school curriculum and just knew immediately that she was the right mentor for me. I mean, Dr. Hair is this, she just has a million ideas about projects. By the end of our first meeting, which was like 30 minutes, she was like, okay, it'd be great to work with you. Here's, you know, 10 ideas for projects just off the top of her head that you could do. And so I was really fortunate to find someone who I think was a good match for my energy levels. I'm someone who likes to be very busy. Dr. Hair does about 30 different things. She's heavily involved in her intestinal rehabilitation program. She's the medical director of our milk bank. She has an R01. She does all this research. She's a guest speaker, a bunch of stuff. And it's all of these diverse interests really you benefit from as a trainee because your faculty is able to bring a lot of different perspectives to projects.

And so really I was very taken with Dr. Hair the first time I met her and knew that she would be a great mentor for me. And I think as well as just having Dr. Hair, she has a very strong support team, which is also very helpful. She has an excellent research coordinator, Laura Gollins, who is one of our dietitians. And it was actually Laura who helped me come up with the idea for this urinary sodium project and suggested that I look into it in the first place. And so not only did I find good mentorship through Dr. Hair directly, I was also able to benefit from her team, her team's involvement and some of the other junior faculty members that she works with, such as Dr. Emily Mierzewski, who's one of our junior faculty over at Texas Children's. And just being able to be part of her team has been such a wonderful experience and has led me to this project and other projects. And I've been very lucky to have that at Baylor.

Srirupa (21:04.1) Yeah, no, that's awesome. And not to mention, I feel like all of us fellows succeed in the fellowship because of our amazing Dr. Hagen, our statistician. And I think that we all owe it to him about all of the support he's given all the last minute. Oh, let me create this table for you before you submit to PAS. He's just amazing. So you have a fantastic team over there. I would like to also kind of highlight that you have beyond your interest in nutrition, you have a lot of interest in medical education and in POCUS. So share with us a little bit about what you've been doing on those two fronts.

Andrew Beverstock (21:40.674) So I was very lucky when I was in the UK to be able to complete a teaching fellowship. So as I mentioned earlier, we do a two year categorical internship in the UK and UK training is much longer than in the US. For pediatrics, it's anywhere from eight to 12 years, depending if you do full time training. And so you spend a lot longer as what used to be called a junior doctor, but has now been officially termed as a resident doctor following after the US terminology. And so when I finished my internship, I decided to take a pediatric teaching job. And there I was able to do lots of simulation and found great mentorship with Dr. Alison Kelly, who was a pediatric rheumatologist in Bristol. And she really got me involved in a lot of QI (Quality Improvement) work then, but also really the primary purpose of that job was to teach the medical students at the University of Bristol on their pediatric clerkship. And so really got involved in medical education early. I think that's one of the unique roles that is available in the UK is these teaching fellowships. And I've been out of the UK for six years now. I don't know whether or not they're still as big as they were, but certainly there were six of us teaching in that group and really loved the education that we were able to provide. So I always knew that I wanted to continue teaching.

All of British medicine is academic medicine and there's learners at every facility. It's very junior led and so there's always involvement at every hospital with medical students and other multidisciplinary learners. And so I knew I wanted to continue that when I came to fellowship. And where I did residency, we were a smaller community hospital on Long Island and we did not have a vascular access team. And so really got involved with POCUS through necessity there. We needed to be able to put IVs in our sick kids and nobody there was ultrasound trained. And I had done some point of care ultrasound for vascular access in the UK, but really was able to develop my skills because we had a machine in our PICU that they use for central line placement, but nobody was using it for PIVs and was able to build my skills there. And then I think as realizing that our residency is short, I realized you have to pass on those skills in order to make sure that all those benefits that you brought don't get lost. And so that is how I got involved in POCUS teaching was initially through teaching peripheral ultrasound guided IVs, which was really a great experience. And so when I came to fellowship, I knew I wanted to stay involved in point of care ultrasounds and Texas Children's had a burgeoning POCUS program at that point. There was some effort prior to the start of my fellowship program in training up fellows and nurse practitioners and attendings in, I think particularly for umbilical line placement and umbilical line assessment, but that had sort of slowed a little bit by the time I got there. We have an excellent summer school where there's a POCUS Day taught by international experts on POCUS, particularly Eric Scheurer, who's one of our PICU docs, who is very well known in the POCUS field. And so I knew that POCUS was something I liked and I wanted to be able to build a skill set in it. And so I spent first year doing as many scans as I could. Anytime a baby had a UVC (Umbilical Venous Catheter) or a UAC (Umbilical Arterial Catheter), I would scan it while I was on service and even come in on research days to do that. And then last year I was lucky enough to be asked to be the POCUS educational champion for our fellowship program. And so as our POCUS program burgeons under the leadership of Dr. Howard Chow, who has really taken things much further at our institution. I've really been able to get involved with that POCUS education, whether it's supervising other fellows doing scans and now getting a little bit involved in some curriculum design to try and build out more regular POCUS teaching rather than limiting it to larger sessions once or twice a year. Because I think that POCUS is a skill that you need to practice to develop, and often you need to carve out some protected time to do that, because if you're not involved in POCUS, I think people can be a little skeptical of how important it is. And so one of the things we're looking to do is introduce weekly teaching sessions and then move into introducing some more formal assessments to assess POCUS competency. Because while it's a great skill to have, you have to be careful that you're doing it under supervision and that you make sure that people are not misinterpreting scans and doing things like pulling back a UVC when in fact the line was not deep and it was the Eustachian valve that you were seeing in the heart. And so trying to introduce a little bit more formal education and some more assessment tools to allow longitudinal assessment rather than just doing it once a year. And so I've been very lucky and I cannot say enough good things about Dr. Chow who really has been an excellent mentor in the POCUS field for me as well.

Srirupa (26:28.92) That's awesome. That's awesome. And you're also part of the junior authors for NeoQuest, which again, I'm also involved in personally, but it is a wonderful way of understanding how question designing works. And this is a very photo based session in NeoReviews where you basically create questions based on pictures that are from the articles that will be published. And I think Andrew's done a fantastic job in the last couple of publications in trying to get some of these questions in the most perfect fashion. And I think that's a skill that any educator needs to have, to create questions, which is fantastic. Well, Andrew, this was great. One last question to you. What is one big advice you'd give to any incoming fellow who's interested in your line of research?

Andrew Beverstock (27:18.926) So I think if you're interested in nutrition research, the thing I would say is think early about how you're going to define your population. What I didn't realize when I started this project is it would be very hard to agree with people on what the definition of faltering growth would be. I thought when you started this project, you say it's easy. We'll take the babies who are growing well and the babies who are not growing well and we'll divide them like that. But there's not a standardized definition in the literature of how you define faltering growth. You use different definitions for babies who are above and below two kilograms. And so I think thinking early and choosing your population very carefully is so important. We ended up using Z-scores and using a Z-score based definition for it, which I think is probably the best approach to use because it accounts for where babies started and where babies end up and the change away from that deviation from normal. But making sure to think about that very carefully and select your population correctly is the thing I really appreciated. That was something we, Dr. Hair and I spent many hours debating and discussing is how to select our population. And so I would say make sure to take your time, slow down, and especially if you're doing a case control study, make sure to select the right population. Because if you don't select that, then the rest of your project is not going to go very far. If you move further on and realize that your babies are in fact not in the group that you thought they were. And I think growth is particularly difficult to define and using some kind of standardized measurement based on Z score would be the best method to do that.

Srirupa (28:52.296) That's amazing. Thank you so much for being on the podcast and sharing your insights and your experiences as you navigated such an important clinical question for us. Thank you again, Andrew, for joining us.

Andrew Beverstock (29:07.768) Thank you for having me.