#294 – Dynamics of human milk expression after very preterm birth in the NICU: Dr Ilana Levene

Hello Friends 👋

In this episode, I had the pleasure of speaking with Dr Ilana Levene, who is now a Neonatology subspeciality trainee at Oxford, England. Il;ana has done some fantastic work on exploring the important topic of human milk expression. She described her randomized control trial in using relaxing techniques to facilitate human milk expression in the NICU. She shared the challenges that she had in conducting her RCT. We also talked about RCTs with negative results and how negative results are also important in conducting research. Ilana has now created a website with printables for parents and staff in the NICU on human milk expression. This can be assessed for free here : http://www.hifn.org/printable . Ilana also shared her interest in perinatal equity and shared details on her project Spectrum which involves gathering photos of the lactating breast conditions/chest from people with a wide spectrum of skin colours. These will be provided as a free educational image library. Currently she is chairing a priority setting partnership for LGBTQIA+ perinatal care. 

----

Notable publications and link to spectrum project

1. Levene, I., Hardy, P., Bell, J.L. et al. Lactation-focused audio relaxation versus standard care for mothers of very preterm infants (the EXPRESS randomised clinical trial). Pediatr Res (2024). https://link.springer.com/article/10.1038/s41390-024-03577-7

2. Levene, I., O'Brien, F., Fewtrell, M. et al. (2024). The relationship of early expressed milk quantity and later full breastmilk feeding after very preterm birth: A cohort study. Maternal & Child Nutrition.https://doi.org/10.1111/mcn.13719

3. Levene, I., Fewtrell, M., Quigley, M. A. et al. (2024). The relationship of milk expression pattern and lactation outcomes after very premature birth: A cohort study. PLOS ONE, 19(7), e0307522.https://doi.org/10.1371/journal.pone.0307522

4. Levene, I., Quigley, M. A., Fewtrell, M. et al. (2024). Does extremely early expression of colostrum after very preterm birth improve mother's own milk quantity? A cohort study. Archives of Disease in Childhood - Fetal and Neonatal Edition, fetalneonatal-2023-326784. https://doi.org/10.1136/archdischild-2023-326784

5. http://www.spectrumlactation.org/ 

6. https://www.hifn.org/printable

   
   

----

Short Bio: Dr Ilana Levene is a neonatal subspecialty trainee in Oxford, England. She recently completed a PhD in the National Perinatal Epidemiology Unit at the University of Oxford. This work examined the dynamics of human milk expression after very preterm birth and how milk yield and breastfeeding outcomes might be improved. She has particularly focused on practical clinical dissemination of the findings, including creating staff and family resources for multiple contexts. These are all available to download at www.hifn.org/printable. Ilana also has an interest in perinatal equity. She is on the steering group for a UK project called Spectrum, which is gathering photos of the lactating breast/chest from people with a wide spectrum of skin colours. These will be provided as a free educational image library. Currently she is chairing a priority setting partnership for LGBTQIA+ perinatal care, which is gathering unanswered research questions on this area and will then collaboratively prioritise these to a "top ten" list.

----

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

Srirupa (00:01.016)

Hello everyone. I hope you all are doing fantastic. Welcome back to the fellows corner of The Incubator podcast. I'm very excited today, mainly because I think that I have accomplished something! The reason I say that is because when I started the fellow series of The Incubator podcast, one of my biggest goals was to get a lot of fellows from all over the world into the show, so we could all talk about our interest in neonatology research.  today we have in our studio Dr. Ilana Levine who is from Oxford, England. She is a neonatology subspecialty trainee from England. I'm so glad that she was able to come onto her show to talk a lot about her research specifically on human milk expression, specifically for advantages in preterm infants. Welcome to our podcast, Ilana. How are you doing today?

Ilana (00:55.906)

Thanks for having me. Yeah, I'm doing very well today.

Srirupa (00:59.18)

Awesome, awesome. I think before we even get started with your interest, I would love to understand how the training typically is in the United Kingdom. If you would like to share that with us, that would be fantastic.

Ilana (01:10.167)

It is very different to America. It takes a lot longer.  we have much  more limited hours as well during our training, which is maybe why it takes a lot longer. So we do medical school. Then we have two years of broad based training, which is called foundation training.  you do various 4-6 month rotations. Then you apply for your specialty; for us that would be pediatrics in general. The program has been shortened very slightly since I started, with eight full years of pediatrics. The last three of those eight years could be a subspecialty, or they could just be general pediatrics, which continues for the whole eight years. So I did five years of full-time equivalent pediatrics, and then applied for Neonatology as a subspecialty. Our hours are limited to 48 hours per week average during that time. Additionally, in the UK, it's very easy to do less than full time working. So I had my first baby in the first year of training of pediatrics. So I've worked part time ever since, and had two full years of maternity leave during that time. So it's taken a very, very long time for me. I started pediatric training in 2012, and I'm still not done.

Srirupa (02:26.008)

I think that is an example of the fact that we're all lifelong learners and no matter how long our subspecialty training goes, I feel like we continue learning. Like it's just a different name, I would say. I'm a first year attending, I'm still learning new things every day. Thank you for sharing that because a lot of us would love to know how training is in other parts of the world. This is great that you're able to share that with us.

So let's delve into your research projects and some amazing things that you have done so far in studying breast milk expression and how that has advantages in the preterm population. So share with us what you've done so far. We'd love to listen from you.

Ilana (03:09.477)

I  stumbled into this area partly because of some personal experiences, partly because of just being on the ward and speaking to parents, and partly because of recognizing that there was a real gap in high quality research in this area. 

We all know on the neonatal unit that the parents often have to do very, very difficult journeys for lactation. The recommendations that we give on how to express milk when your baby's too sick or too immature to breastfeed are super demanding. When we say 8-10 times a day for two to three months; when we say, make sure you wake up in the night one or two times; when we tell them about the volumes we expect people to be able to express and the  stress that comes along with maybe not being able to cover your baby's needs or achieve what you're being asked to do; the stress is immense. As a very evidence minded person, I just wanted to know, what's the evidence behind that those recommendations?

I had exclusively expressed milk for my first baby just because he didn't latch to the breast, not because he was sick.  because he was a term healthy baby, I did actually just follow his pattern. So whenever he fed, I express. So I ended up expressing very much in the way that is recommended because I was mimicking a baby, which is where those recommendations  started. But when I looked at the evidence, I realized that there was very, very little to really support exactly what we're telling people, that it was all based on the  physiology of a normal baby and how they would breastfeed. That is a great starting point. Obviously that's going to be the most likely thing that's going to give success in establishing lactation. But seeing the stress on those parents and seeing how difficult it could be and how many people struggled, I just wanted to find out more about individualized recommendations. Were there some people that we don't have to give such draconian recommendations to? Are there some people that follow those recommendations and still don't manage [to provide enough milk, is] there anything we can do for those? How can we make it easier? 

There is some really good quality research coming out of America, from both nursing-led and doctor-led centers in the US. But in the UK, I think part of the gap is that it's not historically a medical concern. We, as neonatologists, like breast milk, but we don't necessarily see the mum and her breasts as being part of our area. But there isn't a doctor who sees that as their area - the obstetrician is done by then. In the UK, we have a really strong midwifery focus, but the midwives are often also gone after day 10. Neonatal nurses, who do so much of the on-the-ground support, don't necessarily have a really strong research tradition in the UK (although that's starting to change; there's starting to be some investment in trying to help allied health professionals and nurses to become stronger research leaders in the UK). So I saw that it was falling between the silos that we operate in. I also felt that it was quite a feminist issue, that it was part of the female body that just was perhaps ignored by historically male researchers. So for all those reasons, I just said, I've got these questions and I don't think people are answering them satisfactorily, so I'd better try and answer them myself.  that's how I got to my project.

Srirupa (06:31.124)

That's awesome. was just thinking #womanpower. If you have a problem, delve into it yourself. I think that's wonderful that it created a great start. I do understand that, yes, mothers are, in a way, ignored after things happen and they  fall between [specialties and wonder] who do I talk to about this? That's always a question. I think that most women feel that way. Like you said, personal experiences are what teach you what's important in the world of neonatology. That's wonderful that you delved into that. 

I see that you’ve had quite a few publications in this area. Can you share a little bit about some of the work that you've done in investigating the need for a bigger study and what your big study was about?

Ilana (07:18.163)

Over the years, before I even considered a PhD, I was just using small audit projects to ask those questions. I did some small audits where I just spoke to families who were still in the unit and just asked retrospectively, what do you remember about the milk volumes that you got or how early you first expressed, and [looked at] how that was associated with outcomes. I didn't publish those in journals, but I had posters and discussed them with colleagues.

In the UK, we do have the advantage of having a national electronic record. It's not a total electronic record, but there's a summary electronic record that all units in the UK put data into (discharge, admission, and to some extent, daily data). That makes it quite easy to download your own unit's data related to certain things that are measured. I was also able to do some work, which I did publish in journals, looking at the connections between, for example, ethnic background, deprivation and breastfeeding outcomes in our own neonatal unit. But I knew I wanted to do a bigger, more robust project, and that  turned into this PhD idea.

Interestingly, my main questions were about the dynamics of expression and those recommendations (frequency of expression, timing, how much milk you should aim for, etc.). But when I went to somebody who ended up being my PhD supervisor, they said, that's great and those are really interesting questions. But if we're going to have a big group of people who are telling you what they're doing, let's pop a randomized thing in as well. She had done lots of work on relaxation before and said, this is a really simple intervention, it's easy to do and not too complicated for a PhD. So let's do a randomized trial of relaxation in your population. She'd previously supported students to do it and  late preterm babies and term babies in the community. So that turned into that idea. I didn't actually have a particular question about relaxation. But by doing that randomized control trial, I created a cohort of people who also were telling me a lot about their expressing behavior.  that's why I ended up with so many publications from the PhD trial, because before I started, I had lots of questions.  we built that into the design of the RCT to be able to give some answers on the dynamic questions that I was interested in.

Srirupa (09:44.91)

So when you thought about your randomized control trial, I feel like RCTs are probably the highest evidence, but also the toughest to do, especially in the neonatology population, because there's so much emotion involved in the consent process. What challenges did you face while recruiting your mothers for this project?

Ilana (09:59.981)

I was lucky enough that I was doing the PhD within the National Perinatal Epidemiology Unit, which is part of the University of Oxford.  it's both, as it sounds, an epidemiology unit that's related to the full spectrum of perinatal period, but also a clinical trials unit, which has lots of experience in perinatal trials. That was a real bonus for me because I was able to use this relatively easy, small RCT to really learn so much about clinical trials methodology as well. I had the support of tons of amazing staff, statisticians, trial managers, leaders of the clinical trials and steering groups. I had a data monitoring committee, even though the trial was very small and probably didn't really need one, an independent steering group, and that  thing. I had loads of people supporting me with challenges. I had research nurses in various centers who were also randomizing and recruiting people.

We knew actually that it would be a relatively easy study to recruit for, in comparison to lots of neonatal trials because it didn't involve any risk for babies. People didn't have to worry about any  negative impact on their baby. It was all very mother-focused, and it was quite an attractive intervention (the relaxation recording). So from that side of things, we did have quite a high recruitment rates.

The challenge was that we were recruiting people fairly early, so they had to be in the trial before day four after birth. There were worries about talking to people at a stressful time. There were people who just didn't want to speak at all, which was absolutely fine. But also people had perhaps made a decision before they had really experienced the intensity of expressing. Our major challenge was that there were people who joined the trial who seemed very interested and excited and understood what was expected, but then didn't really engage in doing logs of expressing and submitting data. I was able to overcome that a little bit more in my own centre in Oxford, because I was in frequent contact with the participants. In the other centers that I was having research nurses doing the recruitment, they obviously had a lot more studies to deal with and so they weren't necessarily able to go back and back again and talk to people, and trying to maximize the amount of data that we got.

Srirupa (12:21.902)

I think that's understandable for sure. When you're in the center, you have a lot more presence to help in these situations. 

But that's fantastic. Do you mind sharing with us what these relaxation techniques were, and what the findings were on your RCT?

Ilana (12:40.444)

Yep. So many of the relaxation studies that have been done on lactation have all used the same recording as a baseline. This was a recording that was done by a lady called Sheri Menelli, who has a background in meditation and has resources online. I think it was originally available on iTunes for 99 cents, and it was very much designed for working moms who are pumping in America. It was a lactation-specific relaxation and it talked about milk flow, let down, thinking about your baby, etc. Three or four of the studies that have been done have used that study as their baseline. But of course, you have to make changes for the audience that you have. My study particularly needed to make a lot of changes, because the babies themselves were potentially very immature, very sick. The mother would not have breastfed. So this was a very different context.

We did a lot of parent involvement work and there were several strands of things that people were worried about for that recording. One was the implication that things were natural and you could trust your body. In the context of preterm birth, people have had a terribly traumatic event where they definitely could not trust their body. It was definitely not supposed to give birth that early, and it did. So we had to be really careful, because part of the therapeutic quality of the relaxation was potentially getting people to reconnect with their body and trust their body. So we didn't want to cut all of that stuff out, but we had to really think about how we could express it, without making people feel upset and not  clicking with the recording or feeling guilty. So that was one thing. Another thing was that we knew a lot of people would not be happy with their output, their milk yield, and would feel that they were struggling. So we had to think about how we could express the value of milk and saying every drop counts or this is a gift for your baby, but knowing that some people might be really struggling to get any milk at all. [We had to be careful about] referring to breastfeeding or even skin-to-skin, when some babies would not be able to do that. So it was quite challenging to change the initial recording to get something that would work. That was a really difficult moment, because I knew that everything rested on that recording. If we made certain changes, then maybe that was going to mean that it didn't work. And who knows, that may be the case.

The systematic review that I did as part of the PhD on relaxation in general seemed to imply that it doesn't really matter what the content of the relaxation intervention is. It could be music, it could be yoga, it could be mindfulness. So [it probably meant] that didn't necessarily matter as much. In fact, in our study, because we were asking people to listen to the recording frequently, a lot of them said that it was quite boring to listen to the same exact track over and over again. So if people had been able to choose, they probably would have wanted to listen to more of a variety. I did hypothesize that part of the effect of relaxation might be things like picturing letdown, because we know that the visualization of letdown in people with spinal cord injury, for example, is associated with letdown. Even if you don't have the pathways of sensory  connection, visualization of letdown can cause letdown. So I was surprised that that didn't seem to be a particularly strong element.

Srirupa (16:00.782)

What was your final result of your RCT? Did the interventions have an impact on the milk expression at all?

Ilana (16:03.189)

In our study, we did not find that there was any effect on milk quantity, anxiety, post-traumatic stress reactions, or long-term breastfeeding. But people enjoyed listening to the recording, so the people that were in the intervention group said that they really enjoyed listening and that it made them feel relaxed. You could look at these results and say, we haven't really excluded that there is an effect on milk quantity. The other factors like anxiety and post-traumatic stress, there wasn't really much of a signal that there could have been an effect. But in terms of milk yields, if we'd had hundreds more people, it's possible that that same effect estimate that we had could be meaningful. It was about 70 grams more milk in 24 hours, which is about the same as something taking Domperidone to increase milk supply. So that is a possibility that the trial was too small to identify a clinically meaningful effect, but I'm not totally convinced by that. If you adjust for lots of other things, like I did in the exploratory papers, the effect estimates seem to decrease actually. So I think that the relaxation recording that we used did not have a particular impact on milk yield, even though people did tend to listen quite frequently and did enjoy it. So I certainly wouldn't discourage people from using it. But there were people who said that they felt pressured to listen or that they found it boring. So I definitely wouldn't want people in the NICU setting to feel pressured to use it. That's an odd situation to be in, to feel pressured to feel relaxation. It's another thing to feel guilty about that you couldn't relax, that you were too stressed to be able to focus on relaxation. But the systematic review, as I said, did show that overall relaxation seems to be associated with increase in milk quantity. So I think for the general population, we have moderate certainty that that is true, but perhaps for the highest risk populations like we see in the NICU, the effect is going to be much smaller if it's there at all.

Srirupa (18:12.866)

I do agree that there are so many factors that play a role. I think baseline stress for mothers with babies in the NICU is high. I do agree that relaxation is also a very subjective trait, that people want to take in that relaxation technique versus not. It's very individualized. So there might be a lot of variability from that aspect as well. But what's interesting to me is how important a negative result is in research as well. We generally don't talk about that. You're doing your PhD and you probably relate to this comment in a way, because most often we found we won the RCTs with the beautiful interventions that always give the positive results, or we only want studies that give us a really path-breaking results that's positive in a way. But it's also important to talk about results that go against the hypothesis that you had in the first place. What are your thoughts on that?

Ilana (19:13.153)

I think it was quite important from that point of view that I didn't really actually have a stake. I didn't necessarily think the relaxation worked when I started it was something that was a bonus to test. Whereas if you really believed in the intervention, I think it would have been much more psychologically difficult to see the result and decide what to do with it. It's easy when you criticize other people's research to pull it apart and say that didn't have any effect. But when you're inside the study, you start realizing that RCTs are inherently flawed because of the way that people behave: things like not all of the women gave the data; there were imbalances in how many in each group gave logs at different time points; you couldn't necessarily rely on the report of how frequently people were listening. All those kinds of factors made me realize a lot more that the data that we see as gold standard can has flaws. It is really easy to want something to be true, and to pick on the subgroup or the trend that implies that perhaps there is some merit to what's happening.

Srirupa (20:27.404)

I think one of the biggest things that came out of your months and years of research is an amazing tool that contains all kinds of dissemination material for use for milk expression in the NICU. Can you share with us how you created this? Is this something that other NICUs can use? Could you please share that with us?

Ilana (20:28.634)

My driving motivation when I went into the project was to improve the information that we can give to parents. I didn't know whether that was going to be saying the recommendations that we give are totally correct and this is why, or whether it was going to be the recommendations are a bit too harsh and this is something easier, or whether it was going to be this individualization aspect (some people have to express 8-10 times a day, but maybe other people can express more or less frequently). Of course, the study was small. So it's not like I can say that I got the answers, but I got some answers that can be helpful in that individualized support. 

I had three exploratory questions, which I published individual papers on. One was about the timing of first expression. How early you should try and have your first expression after birth? One was the pattern and frequency of expression. One was the milk yield and how that's associated with later breastfeeding outcomes. For each of them, there was data available that could help parents make decisions. At the moment, if you say to somebody express 8-10 times a day, and they say, I can’t, they don't know whether expressing seven times a day is different than expressing four times a day, if you can't express the recommended 8-10 times a day? So I wanted to give people more information about what happens if you can't do the gold standard. What is the absolute impact, [big or small]?

I made a variety of resources: posters that are designed for families to see, say in expressing rooms or in family rooms; leaflets that could be given to parents before giving birth and if they’re at risk of preterm birth, posters for staff (for example, the time to first expression was easier to inform the staff who are going to be supporting mothers, rather than trying to catch families [and show them] a poster in that period). My ultimate idea, that I had right at the beginning, was to produce a pack of postcards that would be attached to breast pumps that would summarize not only my own evidence, but other evidence out there, with an appraisal of all the evidence that we have related to milk expression patterns. I had budget from the NIHR, the National Institute of Health Research in the UK, to do those postcards. 

I've produced all of those things and I've put them together onto a website where people can download and print. That's totally free for everybody to access the link and download PDF and do your own printing. Of course, that can be done globally, but they are in reference to the UK recommendations, which are slightly different to globally. I would love people to use them. You can see the package by going to <www.hifn.org/printable> and the link will be in the notes. All the resources have the University of Oxford National Perinatal Epidemiology Unit and the NIHR information, so people don't need to do anything extra to credit the source. It will be obvious on the information where it comes from.

I would love to hear from you if you do find them useful, because it would be great to get a sense of the impact of the work. That was a big project and the postcards are now available electronically, so you can just access them via QR code, or they could be printed as well. I'm just in process of actually printing them and sending them out to UK and Italy units.

Srirupa (24:24.686)

That's fantastic. For all of our listeners, this website and all of Dr. Levine's work will be part of our episode description today. Please do look at the link. It's a fantastic resource for any information that you would need for milk expression for mothers in the NICU, which I think is a very impactful and fantastic topic in itself.

I heard you say your interest in perinatal equity as well. With the equity aspect of your neonatal research interest, how has that taken your career forward? What are other aspects of your career that you've explored with this perinatal equity interest?

Ilana (25:09.951)

I see that as a separate strand. I started off with this interest in lactation and neonatal nutrition in general. Alongside that, in terms of social conscience, I think like many people in the last 3-5 years or so, it became more obvious that we needed to take action, that it wasn't enough to feel like, racism is bad, or discrimination against LGBTQ people is bad, you had to actually do something in response to that. You had to take your part, especially as a white person and a cisgender heterosexual person. Society is designed to amplify myself and to give me advantage, and therefore I have to use that position of advantage to give back, and make sure that other people are having the resources and being helped to advance themselves. I just continually was asking myself, how can I use the networks that I have and the chances that I've been given to give back in that way. 

One project that came to mind was about images of breasts in the lactation world. So we know, for example, what mastitis looks like in a white breast.  We often talk about the skin being red, for example, as a sign of mastitis. But obviously, the global majority skin color is not necessarily going to look red, it's going to look totally different. That the lack of descriptions and images of people with black and brown skin is going to have a direct impact on the lactation support that we offer. So I took that forward in the UK, got a group of people together who thought that was important, and we formed a group called Spectrum, which is aiming to provide a free image library of lactating breast and chest conditions in people with the whole spectrum of skin colors. We've been working on it for quite a while. There are lots of challenges involved, but the UK lactation community has been really, really supportive. Everybody knows that we need this and some of the big organizations in the UK have given us funding, which is absolutely fantastic. Our current stage is that we're trying to actually get the images, so we have a call out for people to submit images of breast and lactating chest conditions. That is worldwide that people could submit images, so it would be great if people are supporting anybody with lactating conditions, that they could have a look at the Spectrum website and offer to give an image. Once we have sufficient images, we'll obviously launch the site and that will be free to access. Although it will be focused on the UK, it will be globally accessible.

Srirupa (27:57.006)

That's fantastic work.  I see that you're very passionate about the topics that mean a lot to you. That's wonderful that you are delving into aspects of breast milk, lactation, and perinatal equity. That's just fantastic and commendable. We're almost about to end our episode. So one last question that I have for you is, what is one piece of advice would you give to a subspecialty trainee that is looking up to doing breast milk related research and following your pathway?

Ilana (28:29.075)

I think it would just be to get out there and start doing things, because that's how opportunities happen. It's to go to the conferences and contact the people, send emails to people who are doing lactation work somewhere near you, and say that you're interested. That's how things unfold, that you have to give and take, you push a little bit and then an opportunity happens and that raises your profile, then people start thinking of you as the person who's interested in lactation, and things just roll from there. I think also something, which perhaps is more difficult for the American audience, is taking your time. In the UK, it was possible to really take a long time to think about what the questions were and write a high quality application that was able to get funding, and that pays off so that you can do the highest quality research, rather than just any research.

Srirupa (29:25.358)

That's fantastic advice.  I'm pretty sure you will be inspiring a lot of the subspecialty trainees around the globe to follow your pathway. Thank you so much for being on our podcast today. It was a delight having you on and talking a lot about your interest and your passions. Thank you so much.

Ilana (29:43.342)

Thanks for having me.