#313 - 📑 Journal Club - The Complete Episode from May 25th 2025

Hello friends 👋

In this week’s Journal Club, Ben and Daphna unpack a wide range of recent neonatal studies with pragmatic, practice-centered discussion. First, they explore a study on low-dose dexamethasone for BPD in preterm infants, showing potential benefits in brain development and motor outcomes—despite ongoing concerns about long-term effects. Next, they discuss a large dataset analysis of oxygen and respiratory support trajectories in extremely preterm infants, offering real-world FiO2 trends and benchmarks that may help frame clinical decisions and counseling.

They also examine the diagnostic limits of consumer-grade pulse oximeters, like the Owlet, comparing their accuracy to hospital-grade monitors—raising real concerns about missed events. A safe sleep initiative study offers evidence that modeling and education during birth hospitalization can improve post-discharge sleep practices, especially across different demographic groups. Finally, they review parent engagement with NICU-focused online health communities, identifying both the benefits and potential friction these platforms create in team-family communication.

From cerebral oxygenation during kangaroo care to the use of enemas in ELBW infants, this episode covers it all—with a focus on what clinicians can take back to the bedside.

Listen in and join the conversation. 

----

The articles covered on today’s episode of the podcast can be found here 👇

Trajectory of Postnatal Oxygen Requirement in Extremely Preterm Infants.

Groves AM, Bennett MM, Loyd J, Clark RH, Tolia VN.J Pediatr. 2025 Feb;277:114414. doi: 10.1016/j.jpeds.2024.114414. Epub 2024 Nov 20.PMID: 39577761

Postnatal dexamethasone treatment for preterm infants at high risk for bronchopulmonary dysplasia is associated with improved regional brain volumes: a prospective cohort study.

Chandwani R, Kline J, Altaye M, Parikh N.Arch Dis Child Fetal Neonatal Ed. 2025 May 13:fetalneonatal-2024-328438. doi: 10.1136/archdischild-2024-328438. Online ahead of print.PMID: 40360237

Diagnostic accuracy of an over-the-counter infant pulse oximeter for cardiorespiratory events.

Travers CP, Nakhmani A, Armstead KM, Benz RL, Foshee KM, Carlo WA.Arch Dis Child Fetal Neonatal Ed. 2025 May 12:fetalneonatal-2025-328540. doi: 10.1136/archdischild-2025-328540. Online ahead of print.PMID: 40355254

A Hospital-Based Initiative for Infant Safe Sleep Practice.

Decker CM, Dunlevey E, Nguyen L, Stence KJ, McCarty E, Jean-Charles TG, Trego T, Ma ZQ.Pediatrics. 2025 Feb 1;155(2):e2024067659. doi: 10.1542/peds.2024-067659.PMID: 39799962

Neonatologists' perceptions of uncertainty: a national survey.

Rholl E, Krick JA, Leuthner SR, Pan AY, Challa SA, Kukora S.J Perinatol. 2025 Apr 19. doi: 10.1038/s41372-025-02292-3. Online ahead of print.PMID: 40253559 No abstract available.

The effect of maternal position on cerebral oxygenation in premature infants during Kangaroo care: a randomised controlled trial.

Stapleton I, Murphy S, Vaughan S, Walsh BH, Natchimuthu K, Livingstone V, Dempsey E.J Perinatol. 2025 Apr 5. doi: 10.1038/s41372-025-02287-0. Online ahead of print.PMID: 40186001

Hydrostatic low-volume enemas in infants with birth weight ≤1000 g or gestational age ≤28 weeks: A controlled interventional study.

Stock T, Kamp AM, Waitz M, Riedl-Seifert T, Jenke AC.J Pediatr Gastroenterol Nutr. 2025 May 8. doi: 10.1002/jpn3.70055. Online ahead of print.PMID: 40344423

The new normal: parental use of online health communities in the NICU.

Adams SY, Tucker R, Lechner BE.Pediatr Res. 2022 Jun;91(7):1827-1833. doi: 10.1038/s41390-021-01684-3. Epub 2021 Aug 17.PMID: 34404928

----

Watch this week's Journal Club on YouTube 👇

----

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

Ben:Hello everybody, welcome back to the Incubator Podcast. We are back today for an episode of Journal Club. Daphna, how's it going?

Daphna:I'm well. I have some…dust or crumbs here. But I'm good.

Ben:Wait, what are you talking about—dust on your desk?

Daphna:There’s crumbs on my desk! I must have been having a snack the last time I was here.

Ben:Oh, I see. You're not talking about literal crumbs. I was like, why and how is that compelling for the audience to know?

Daphna:It's not. But speaking of my desk, I'm in my new little studio here for all our video productions.

Ben:The audio episodes of the podcast have always been on YouTube, somehow. I think our team used to put out an audio-only format. Now, there's more video content coming out.

Daphna:So if you really want to look at us while we’re talking.

Ben:I know some hospitals have big screens in the lobby. So if you want to put us up there, as patients are entering the building...

Daphna:For sure. Or, our Beyond the Beeps colleagues. You know, it’d be nice to be on there at every entryway.

Ben:No, no—me. Specifically me. Even if you have to crop everything, it just needs to be me.

Daphna:Oh gosh.

Ben:That was a joke, please. Even the release of these episodes on YouTube is a big ordeal for me to accept. Anyway, we have a very nice episode of Journal Club. We're not going to waste too much of your time with our usual banter. Daphna, anything we should mention? Obviously, the Delphi Conference is coming along. We’ll have a special video episode coming up very soon. Our agenda is mostly finalized.

Daphna:Yeah. And people don’t have to wait—they could be registering now, making their plans. Our new hotel is close to the other one, if you came last year, but it’s really cool. It’s a very South Florida vibe. We think people will really enjoy it. There’s space if you want to bring your families or loved ones.

Ben:Yeah, I think the vision for the conference is crystallizing more and more. We've done something fairly consistent for the past two editions, but now it’s becoming more obvious that we want this to feel like a retreat—something people can go to for a break from the grind. A time to recoup, recharge—all while being entertained by lectures and conversations at a very high level, with top-tier speakers and content. So yeah, think of it that way. The hope is that you learn a lot, feel refreshed, and go back to your institution energized and full of ideas. My goal is that someone says, “What’s gotten into you?” when you return.

Dapna:It's that Delphi glow, you know? And we should remind people about the giveaway for our survey.

Ben:That's right, we’re turning four! We’re toddlers now. We’re officially PICU material. We've adopted this new modality of giveaways where every year on our anniversary, May 4th, we put out a short survey to gather feedback from the community. It's a great way for us to get that feedback, and it's super simple. I’ve done the form myself—not to enter the giveaway, but just to see how long it takes. It literally takes three minutes. And we have tons of cool gifts: a laptop, headphones, speakers, merchandise from the Incubator shop, tickets for the conference, books—it’s not just one big prize. There are lots of prizes, so hopefully something goes your way. And we leave the giveaway open for a while so everyone has time. I mean, how many times have you received a DocuSign, and by the time you open it, it's expired?

Daphna:Basically every time I've ever received a DocuSign.

Ben:Yeah. So go check it out. It's on our webpage, our website, and our social media. We've tried to publicize it as much as possible. We want to make sure everybody gets a chance to participate.

Alright—five minutes into the episode, let’s begin. We’re reviewing the latest evidence in peer-reviewed publications in neonatology. And if it’s okay with you, I’ll start with an article on postnatal steroids. The article I found is from Archives of Disease in Childhood and it's titled “Postnatal Dexamethasone Treatment for Preterm Infants at High Risk of Bronchopulmonary Dysplasia is Associated with Improved Regional Brain Volumes.” Definitely a spoiler in the title—but a good one. It made me want to read this paper.

It’s always interesting to revisit the history of postnatal corticosteroids. We know they’ve been used for the prevention and treatment of BPD in at-risk infants. However, the therapy has always been controversial due to concerns about adverse effects on the developing brain—including a potential increase in the risk of cerebral palsy and mortality. That’s why, in the past, clinicians have been advised against using postnatal corticosteroids—especially high doses early on, within the first week of life, for prolonged periods, and even in babies with relatively low risk of BPD.

One of the landmark papers in this discussion was the DART study, which found that giving very low-dose dexamethasone after the first week of life to ventilator-dependent, extremely low birth weight infants at high risk of BPD reduced the duration of intubation and did not increase short-term complications, neurodevelopmental impairment, or death at two years corrected age. Later meta-analyses suggested that when corticosteroids are given only to infants with a baseline BPD risk (around 50% or higher) and not in the first week, the treatment provides a net benefit, including reductions in cerebral palsy and death. Despite this, we still face challenges when discussing steroid use with families. The risk-benefit conversation is tough, especially when a simple internet search may alarm parents about potential brain effects, without the context we just discussed.

The research group in question, out of Cincinnati, had previously reported that dexamethasone given via the DART protocol for severe BPD was associated with decreased risk of diffuse white matter abnormalities on T2-weighted MRI. This is a known predictor of adverse neurodevelopmental outcomes. But that particular study didn’t examine other sensitive or objective measures of early brain injury, brain maturation, or long-term neurodevelopmental outcomes. So, the objective of this new study was to evaluate how low-dose dexamethasone therapy for BPD affects brain volumes at term-equivalent age and motor outcomes at two years corrected age in a regional cohort of preterm infants.

This was a prospective study that enrolled a cohort of about 392 preterm infants, born at or before 32 weeks’ gestation, between September 2016 and November 2019. They were recruited from five level III and IV NICUs as part of the Cincinnati Infant Neurodevelopmental Early Prediction Study, also known as CINEPS. The infants who were excluded from the study had chromosomal or congenital anomalies affecting the brain, spine, or heart. BPD was defined based on the type of respiratory support provided at 36 weeks postmenstrual age. Out of this large cohort, almost 400 infants, only 41 were actually treated with dexamethasone according to the DART protocol. So, these are the babies the study focuses on. It's a relatively small cohort, so that’s something to keep in mind as we look at the results.

The motor outcomes in this study were assessed using the Bayley Scales of Infant and Toddler Development, Third Edition. It's important to note that the children in this prospective study were not randomized. They weren’t randomly assigned to receive DART or not. So, to address potential selection bias and better understand the independent effects of dexamethasone, the researchers used a propensity scoring approach. This helped balance the treated and untreated groups based on baseline demographic and clinical factors. Essentially, they created a matched comparison group of babies who did not receive DART.

Let’s get into some of the results. The 41 infants who received dexamethasone were treated sometime between 32 and 40 weeks postmenstrual age. Their mean gestational age at birth was 25.5 weeks, with a standard deviation of 1.6 weeks. The mean birth weight was 800 grams, with a standard deviation of 229 grams. Among the 41 infants treated with dexamethasone, 81% developed severe BPD. All 41 infants developed BPD and were treated with a low cumulative dose of dexamethasone under the DART regimen—0.9 mg/kg administered orally or via injection over 10 days after the first postnatal week. It raises a bit of a chicken-and-egg question: if you’re at high risk for BPD and you’re being given dexamethasone, it may not be surprising that you end up developing BPD. Still, it's depressing when the goal of DART is to prevent BPD and yet 100% of the infants developed it.

In terms of outcomes, on multiple linear regression analysis controlling for additional confounders, dexamethasone was significantly associated with a larger cerebellar volume (difference of 0.5; p = 0.021). It was also linked to a significantly larger subcortical gray matter volume (difference of 0.138; p = 0.03) relative to total brain volume. Initial Bayley-3 motor scores were significantly lower in the dexamethasone group (p < 0.001), but when adjusted for covariates, that result reversed—dexamethasone was then positively correlated with higher Bayley-3 motor scores (p = 0.02). As for cortical surface area, there was no significant association with dexamethasone therapy (p = 0.3 after multivariable weighted linear regression).

These are very interesting results, especially when counseling families about the use of postnatal corticosteroids. In summary, the authors conclude that dexamethasone treatment for BPD prevention is associated with larger cerebellar and subcortical gray matter volumes at term-equivalent age and improved motor development at two years of age in this cohort of preterm infants. Given the lack of alternative therapies to facilitate extubation, the study underscores the importance of considering dexamethasone after the first postnatal week to help reduce the risk of adverse neurodevelopmental sequelae from BPD in high-risk infants.

Daphna:Yeah, I agree with you. I think the takeaway is just right. We don’t have a lot to offer, aside from nutrition and developmental care. We're really doing the fine-tuning when it comes to what our medications can do. So yeah, thank you for sharing that one.

I have a paper in a similar vein, from the Journal of Pediatrics. It’s titled “Trajectory of Postnatal Oxygen Requirement in Extremely Preterm Infants.” The lead author is Alan Groves, senior author Veeral Tolia. What they wanted to do was describe trends in oxygen requirements and modes of respiratory support in extremely preterm infants during the first three months after birth, broken down by gestational age.

Previous studies have shown this kind of biphasic pattern in respiratory support: infants need a lot of oxygen on day one, then that requirement drops (the so-called "honeymoon period") and then increases again in a second peak. This study wanted to dig deeper into that. They used the Pediatrix Medical Group database and included infants discharged between January 1, 2016, and December 31, 2021. Infants were eligible if they were born between 22+0 weeks and 27+6 weeks. They were included if inborn or transferred into the NICU on day 0 or 1 of life. Major congenital anomalies were excluded.

For each infant, they looked at daily respiratory support documented in progress notes and recorded daily FiO₂. For babies on low-flow nasal cannula, they assigned an FiO₂ of 0.25. Babies on room air with no support were assigned 0.21. They noted that during this time period, many papers had come out about targeting O₂ saturations between 90–95%, though they didn’t account for individual unit policies or altitude—which, if you’re studying for boards, you need to know why altitude could impact FiO₂ needs!

Ben:Oh my god, I’ve been in some quality meetings, and it’s such a headache for centers at high altitudes and their BPD rates. Do you apply correction? It’s really complicated.

Daphna:Very complicated, right? I don’t think anyone has a clear answer yet. So, they created FiO₂ trajectory curves stratified by gestational age in completed weeks. Then they mapped the 10th, 25th, 50th, 75th, and 90th percentiles of daily FiO₂ over 84 postnatal days. They also looked at support types by gestational age through week 11 and calculated BPD rates based on infants needing oxygen or support at 36 weeks, based on 2018 NIH definitions. The denominator included only infants who survived to 36 weeks and had complete respiratory data.

They included 16,386 infants from 221 NICUs. The median gestational age and birth weight were 25 weeks and 765 grams. About 23% (3,800 infants) died before discharge. They had respiratory support data for nearly 14,000 infants, and 43% met BPD criteria.

A few highlights: the graphs in this paper are phenomenal. They are something you could print out and use for nursing staff, learners, or even with parents. They had 600 22-weekers, almost 2000 23-weekers, 2800 24-weekers, 3200 25-weekers, 3600 26-weekers, and just over 4000 27-weekers. So quiet a varied group, but a nice proportion of small babies.

For 22-weekers, the median FiO₂ followed a clear biphasic pattern, with a peak on day 1 (~0.5 FiO₂), dropping to a nadir of median FiO2 0.3 by day 4, then rising again to about 0.45 around days 12–20. After that, in the 2nd to 3rd week, the FiO₂ gradually decreased through day 84. This pattern was consistent across gestational ages, but the median FiO₂ values were lower in older gestational age infants. So on the graph of postnatal days vs. median FiO₂: 22-weekers peaked around 0.45, 23-weekers around 0.38, 24-weekers at 0.33, 25-weekers at 0.30, 26-weekers at 0.25, and 27-weekers between 0.21–0.24. You can clearly see the stepwise pattern, and it becomes smoother after day 20-24 or so in all of the groups.

For respiratory support, in 22-weekers, high-frequency and conventional ventilation were predominant for the first four weeks. After that, there was a gradual increase in nasal prong ventilation, CPAP, and high-flow nasal cannula over the next two months. Very few of these infants were weaned off respiratory support by 34 weeks corrected age. In contrast, babies born at higher gestations showed earlier use of nasal prongs, CPAP, and high flow; and fewer needed mechanical ventilation. The visual tables are great, showing support types over time by gestational age. It’s exactly what you’d expect: more intensive support in younger babies, less in older ones.

Basically, the take-home is that there's this biphasic pattern occurring independent of gestational age at birth, though the FiO2 peaks differ. So, they have this early oxygen requirement—maybe they get surfactant, then it falls to a nadir around days three to seven, and there's a second peak at 14 to 21 days. This pattern is seen in all gestational age groups.

I think it’s really neat. You know, why is this useful? Sometimes we feel, “Gosh, this baby was getting better and now is getting worse. Do I work this baby up? Is this a normal transition?” For example, an increase in oxygen requirement at day 30 would be more concerning clinically than one at day 10, when we can expect babies to have their peak. Certainly, these trajectory curves could help us in research, in clinically relevant outcomes, and in counseling of both staff and parents. So I thought this was neat. I hope people will take a look at the curves.

Ben:Yeah, I think it’s always very valuable when we can take close to 20,000 kids and actually have that data. I really appreciate the paper. Also, by the way, I’ve now been traumatized by scientific papers. When I open a paper, I’m like, please make sure there are graphs…if there aren’t, I will lose it!

Daphna:And I’ll say, this paper was mostly graphs, so loved it.

Ben:Yeah, sometimes you see a title and you think, “This is going to have the graph I’m excited to see,” and then you look through and it’s just tables, tables. Like, my God, I just want to see your graph. That being said, I think it’s a fascinating figure. It’s sort of a mirror image for all the different gestational ages staggered over higher FiO2s. I think that’s very interesting. I’m going to say the usefulness of that particular graph probably, to me, stops at about a month of age. I think in the first couple of weeks after birth, there’s little variability in how we handle these patients; we all practice fairly similarly. But after that, there are many differences, so it’s hard to know if this is broadly applicable. You can see that toward the tail end of the curve, pretty much all the different gestational ages, except 27 weeks, level off at around 25% FiO2. Is that real? I’ve seen 26-weekers on room air at 36 weeks, so there are probably differences contributing to this outcome. This data is helpful because it gives a broad overview of what many units are doing. And it could definitely be used when counseling families about what to expect and what’s a reasonable outcome. But there are definitely differences between units and patients, especially toward the tail end of this graph.

Daphna:I also thought, if you scroll to figure two, which is graphs by percentiles, it just shows how variable these 22-weekers are the most variable of the cohort. I think that’s interesting too: some babies really do “well” from the beginning and some really do “poorly.” They mostly end up in the same place, though.

Ben:Exactly. I think this is where this is very helpful. You bring this to a prenatal consult, talk about antenatal steroids, and say, “I’m hoping the steroids help shift your baby from a higher risk group to a lower risk one.” Parents can visualize this better, because it’s very esoteric when we talk to them during prenatal consults.

Okay, I’m next. My next paper is from the Archives of Diseases in Childhood again, two articles back to back. This one is from Alabama and is called “Diagnostic Accuracy of an Over-the-counter Infant Pulse Oximeter for Cardio-Respiratory Events.” Very high yield, I think. I’ll tell you afterwards why I have a particular interest in that paper, but the consumer baby monitor market is booming. There’s a growing number of over-the-counter smartphone-integrated infant monitors, driven by understandable parental anxiety about recognizing illness or preventing sudden unexpected infant death (SUID).

But the American Academy of Pediatrics recommends against routine use of home monitors. Why? Because there isn’t enough evidence these devices reduce morbidity or mortality. More critically, the diagnostic accuracy of these over-the-counter monitors compared to hospital-grade equipment hasn’t been validated. Until recently, none of these devices were FDA-cleared to diagnose, treat, or prevent disease. The concern is if these devices are inaccurate, they might give parents a false sense of security or even encourage unsafe sleep practices. Previous work testing two such monitors found only one had decent accuracy, underscoring the importance of rigorous clinical testing.

So this study aimed to test the diagnostic accuracy of an over-the-counter infant pulse oximeter, the Owlet OSS 3.0. I’ve discharged parents on home monitors, and the ones from our DMEs look awful, like they were made in the fifties and not the most aesthetic.

Daphna:That’s right. Parents are like, “This is the updated tech for my 2025 baby?”

Ben:How many times have parents just pulled up Amazon and asked, “Can I just get the sock?” I think there was a point where the FDA had removed approval, but now the Owlet is back in the FDA’s good graces. It’s hard for us to say yes or no—can we say, sure, use that? Parents tell me, “I’ll pay for it; don’t worry about insurance. I want this one.” It’s about $250 to $300. I don’t know if it’s safe or if I can in good conscience say yes, it’s equivalent. This paper was a long-awaited study.

It was a prospective single-arm observational study at the University of Alabama. Infants were eligible if they were under 44 weeks post-menstrual age, weighed at least 1,500 grams, and were off ventilatory support and phototherapy for at least 48 hours. Major malformations, neuromuscular conditions, skin infections, terminal illness, or limited care decisions were exclusion criteria.

There’s an important technical detail: the study examined two types of output from the pulse oximeter—labeled as raw data versus smoothed data. Raw data refers to unprocessed readings every 10 seconds, with minimal filtering for movement or signal interference. Smoothed data is processed through a proprietary algorithm that dampens sudden changes, filters out motion artifacts, and suppresses noisy readings that don’t meet quality thresholds, aiming to reduce false alarms. This, however, may miss real but brief events.

After enrollment, each infant was hooked up to hospital-grade EKG and pulse oximetry, concurrently with the Owlet device, for 48 hours. They followed the Owlet’s instructions rigorously, checking every couple of hours. The primary goals were to assess if the device could detect bradycardia (heart rate below 50 bpm for at least 3 seconds) and hypoxemia (oxygen saturation less than 80% for at least 3 seconds). Secondary goals explored performance under other thresholds (mild hypoxemia, higher heart rate limits, longer episode durations).

Seventy-five infants were enrolled between April and July 2023; data from 66 infants were included after excluding corrupted data. About 20% were on supplemental oxygen. Average gestational age was about 30+4 weeks, and they were enrolled around 35 weeks post-menstrual age.

Let’s talk diagnostic performance. When I read the paper post-call, they discussed sensitivity and specificity, concepts I’m familiar with, but I thought I’d review what they mean in this context, in case you’re also post-call and listening. Sensitivity is the ability of the device to detect true events—if a bradycardia or hypoxemia episode actually happens, how often does the device catch it? It’s about detection. Specificity tells us how often the device correctly identifies no event, meaning no false alarms.

For detecting heart rate below 50 bpm for at least 3 seconds, specificity was 100% for smoothed data and 99% for raw data. Sensitivity was just 6% for smoothed data and 39% for raw data. For saturation less than 80% for at least 3 seconds, specificity was again high—100% for smoothed data and 96% for raw data. Sensitivity was only 14% for smoothed data, but much better for raw data at 74%. That difference reflects the proprietary algorithm’s effect: trying to reduce false alarms may come at the cost of missing real events. So when the device flagged an event, it was almost always correct, but it missed many events, especially with the smoothed algorithm.

The pulse rate output of the smoothed device strongly correlated with hospital heart rate, with an average error of just 2.7 bpm. Sensitivity for bradycardia detection varied widely from 0 to 100% depending on episode duration and severity. Saturation correlation was moderate (R² = 0.48) with an average error of 2.5 points. Sensitivity for detecting hypoxemia improved with raw data, longer episodes, or higher oxygen thresholds. Specificity stayed high across definitions, but sensitivity was generally higher for raw signals and more severe or longer events.

To wrap up, this straightforward but valuable single-center prospective study shows that in infants at high risk of cardiopulmonary events, this over-the-counter pulse oximeter had excellent specificity but low sensitivity for short bradycardia and hypoxemia episodes. That means it rarely gave false alarms but often missed brief events unless they were longer or more severe, unless the raw data was used. These results underscore that while consumer-grade pulse oximeters like the Owlet might be helpful in certain conditions, they are not reliable enough, at least in their current form, for catching brief but potentially critical events. The authors conclude — and I’m sure this is an open door to all the other partners in the industry — that further trials are needed to see how this can impact real-world outcomes in healthcare use. So I thought this was very interesting.

Daphna:Yeah, you know, because I think in general we tell parents there are a lot of false alarms, right? That it’s very annoying, it’s going to alarm all the time, this causes a lot of stress for families. And that’s actually not true. I mean, it probably does alarm when the baby’s moving — things like the same things we see with our monitors in the NICU. But now I think we can really give parents the information.

Ben:Would you rather have something that’s going to alarm a little bit more but is reliable, versus something that alarms less but might miss something terrifying? That would cause me more anxiety than anything else.

Daphna:Yeah, totally agree. For sure. And that concept of feeling like, okay, it’s safe — like, I’ll know if something’s going on. Or maybe I can choose less safe sleeping choices because I’m on a monitor. Things like that.

Ben:And maybe you can even quantify which baby this is good for. If you have a baby with extremely low risk where the parents are very anxious, maybe the Owlet is a great option because the risk is extremely low. But a baby who goes home on oxygen, with a history of extremely low birth weight and BPD? No, that’s not going to do it. My answer would be: not for you. Not yet, at least.

Daphna:Man, Owlets. We get asked about them every single day.

Ben:Because what we have to offer for DMEs looks antiquated.

Daphna:That’s true. But I mean, sometimes we don’t even have the car seat, but they’ve got the Owlet.

Ben:Yeah. And I feel like we’re going to have a Tech Tuesday episode with Dr. Brezinski from Harvard about a phototherapy device she came up with. We’ll talk about some of the benefits of this particular device. It’s called the bili-hut from Little Sparrows. It looks beautiful! And that’s what parents want, especially in this era of “designer babies” where everything the baby wears, from the stroller to the pajamas, is beautifully designed. You bring out our equipment and it looks just so bad.

Daphna:I know, I know, the social media era.

Since we’re talking about safe sleep. It’s called “A Hospital-Based Initiative for Infant Safe Sleep Practice.” Lead author Caryn Decker, senior author Zhen-Qiang Ma. And this is coming in Pediatrics. So I’ll say this was done in the newborn nursery on well babies, but the point is well taken. They basically wanted to see if a hospital-based initiative of education and behavior modeling inpatients is associated with increased likelihood of safe infant sleep practices at home.

We know babies born early or with medical complexity are at even higher risk for sudden unexpected infant death… SUID, which I don’t think people even say. I think they say S-U-I-D, which is way harder to say than SIDS.

Ben:Yeah, like when they changed ALTE to BRUE, I can get with that. I can still say that, but it has to be equivalent in terms of pronunciation for foreign speakers or English-challenged people like me.

Daphna:So in 2016, the Pennsylvania Department of Health funded this initiative about safe sleep. It included infant safe sleep policies, nursing staff education, parent education through posters, videos, brochures. Importantly, nursing staff modeled and reinforced infant safe sleep by placing the infant to sleep in the in-room bassinet on their back without additional objects for the duration of the birth hospitalization.

Pennsylvania also has this PRAMS system — Pregnancy Risk Assessment Monitoring System — a survey collected by maternal self-report on infant safe sleep practices. They looked at a retrospective population-based study using these linked data sets, comparing mother-infant dyads exposed to the safe sleep initiative versus those who were not. The survey assessed these practices: (1) mother reports most often placing infant to sleep on back and not side, stomach, or combination, (2) Always placing infant on a separate approved sleep surface such as crib, bassinet, pack and play, (3) Usually placing infant without soft bedding or objects like toys and pillows, and (4) Usually placing infant alone in own crib in the same room.

They examined the association between exposure to the initiative and these four self-reported safe sleep practices. They also considered other factors associated with sleep practice: year of birth, maternal race/ethnicity, education, age, parity, marital status, insurance, smoking, provider advice, prenatal care initiation, prematurity, and breastfeeding.

Between 2017 and 2021, 637 respondents were exposed to the Safe Sleep Initiative, and 4,861 control or unexposed. Among PRAMS respondents exposed to the Safe Sleep Initiative, the prevalence of reporting each of the four infant safe sleep practices increased between 2017 and 2021: back sleep from 4% to nearly 40%, separate surface from 3% to 43%, no soft bedding from 3% to 30%, and room sharing without bed sharing from 4% to 37%.

Infant safe sleep behavior differed by maternal race in both groups, but notably, a higher proportion of Black respondents exposed to the initiative reported each safe sleep behavior compared with unexposed Black respondents. A similar pattern was seen among Hispanic respondents. However, the discussion notes that, after adjustment, Black respondents were less likely to report placing their infant on their back compared with White respondents and were less likely to report multiple safe sleep practices when in a cumulative model. This shows that there’s still work to do, especially for different subgroups of families.

More than 90% of respondents who placed infants on their back on a separate surface or without soft bedding said they were told by a healthcare worker about the practice after pregnancy, consistent across groups. Fewer respondents who reported room sharing without bed sharing indicated they had been told about this practice. Minimal differences were observed by maternal education, but infant safe sleep behaviors were less common among respondents aged 19 or younger or 40 and older in both groups.

Eighty percent of respondents practicing safe sleep behaviors in both groups had initiated breastfeeding. A slightly higher proportion of those exposed to the initiative who reported placing infants on their back, separate surface, or without soft bedding were still breastfeeding at survey time compared to unexposed respondents. Similarly, more exposed respondents who were room sharing without bed sharing were still breastfeeding (56% vs. 52%).

Most respondents who reported infant safe sleep were married, had a prior birth, had term births, had initiated prenatal care, were not smoking at the time of the survey, and had private insurance. Smoking at the time of the survey was associated with decreased likelihood of placing infants on separate surfaces, without soft bedding, or room sharing without bed sharing. Respondents with Medicaid or other non-private insurance were less likely to report back sleep or sleep without soft bedding. Unmarried respondents were less likely to report sleep on a separate surface.

In a logistic regression model, PRAMS respondents exposed to the safe sleep initiative had a 72% increase in odds of placing infants on their backs, 48% increase in always placing infants on a separate sleep surface, and 41% increase in usually placing infants without soft bedding, compared with unexposed respondents. There was no significant association between exposure and room sharing without bed sharing. The logistic regression also suggested respondents exposed to the initiative were 38% more likely to practice more recommended safe sleep practices than those unexposed.

I think the data is not surprising, but we do a lot of unsafe sleep practices in the NICU because we have to and because we can, since the babies are on monitoring. This group, in the weeks leading up to discharge, is even more important. We have to model for parents that their babies can sleep in these recommended ways at home. I’ve worked in lots of units, even those focused on safe sleep, and we still miss some babies. Certain babies we put on their bellies because we say they sleep or feed better. But modeling and education are really important, so I plan to spend more time at the bedside with those lower-level babies.

Ben:Absolutely agree. It’s a shame babies who went to the NICU were considered unexposed. Obviously, we can’t put the burden of safe sleep on critically ill newborns, but closer to discharge, parents start to emulate what nurses and providers do. In anticipation of going home, that’s when safe sleep practices kick in, not putting babies on their bellies for a good night’s sleep just because they’re on a monitor. I completely agree. Interesting to see.

Daphna:We have to set them up for success. If babies don’t learn this with us, how will they sleep that way at home?

Ben:Absolutely. We’re running short on time, so I want to mention two quick papers. The first, from the Journal of Perinatology, is called “The Effect of Maternal Positioning on Cerebral Oxygenation in Premature Infants During Kangaroo Care: A Randomized Control Trial.” This study from Ireland aimed to assess whether an optimal maternal position — reclining at 30° versus 60° during kangaroo care — affects cerebral oxygenation. Using near-infrared spectroscopy on 20 infants (median 28 weeks gestation), they showed position did not impact cerebral oxygenation. Interesting paper, but not worth much more time.

The last paper, also from Journal of Perinatology, is “Neonatologists' Perception of Uncertainty: A National Survey.” Limited research has looked at how neonatologists experience uncertainty beyond communication with parents. In other specialties, intolerance of uncertainty correlates with burnout. This is a big deal in neonatology, where high-stakes decisions and ambiguity are part of our daily routine. The goal was to describe neonatologists' reaction to uncertainty in patient care.

They used the revised Physician’s Reactions to Uncertainty scale (PRU), which captures anxiety, concern about bad outcomes, reluctance to disclose uncertainty to patients, and reluctance to disclose uncertainty to physicians. Lower scores mean greater tolerance to uncertainty. The survey was distributed via the AAP Section on Neonatal-Perinatal Medicine listserv between March and April 2023. A total of 345 neonatologists responded, spanning career stages. Thirty-eight percent had practiced more than 20 years, 64% were female, 70% identified as white, and 65% worked in academic centers.

When analyzing PRU scores, there was a significant difference by years in practice: more experienced clinicians had lower scores, meaning greater tolerance for uncertainty. The difference was specifically in anxiety and concerns about bad outcomes that showed differences.

What does this all mean? The study suggests that uncertainty is a significant concern for neonatologists, especially early in our careers. The authors are basically using these results to say we need to explore this further not just for our own well-being, but so that we can offer confident, compassionate support to families navigating the NICU experience. It's one of those areas where culture, mentorship, and self-awareness all come into play. This study is a strong first step in understanding how this all unfolds. I thought it was very interesting.

Daphna:Yeah, it certainly impacts how we individually counsel parents. If you’ve got people coming on and off and parents hear different things, I think that can be stressful for everyone, the team and the families, and definitely impacts satisfaction with our communication.

Ben:Yeah. They have this beautiful figure in the paper. It's a very short communication in the Journal of Perinatology where they have bar graphs looking at different aspects of that scale: diagnostic uncertainty, prognostic uncertainty, evaluation and management uncertainty, uncertainty around what to offer, and uncertainty around limit setting. The X-axis shows years in practice: fellow, less than 5 years, 5 to 10, 11 to 20, or more than 20 years. It’s interesting to see basically how there’s a shift in agreement over the years of practice. I’ve seen this consistently across my career. Especially early on as a fellow, I would be anxious about every single aspect. If a gas had a pH of 7.2, I would worry the whole night. Then you have some attendings close to retirement who are so chill about everything. I think this has to do with years of experience and all the cases you’ve seen. It feels logical because early in your career, you haven’t seen everything, you haven’t seen the variability nature has to offer. So you agonize, "Am I doing something wrong?"

Daphna:I agree. This figure is interesting because I thought maybe with different types of uncertainty, we wouldn’t see such a difference. For example, evaluation and management, or diagnostic uncertainty. But these different graphs, regardless of type, are almost copies of one another. The more years in practice, the more comfortable you are with uncertainty across the board, which I thought was interesting. I expected more variability in prognostics, limit setting, and what to offer, as compared to diagnostic or evaluation management, but we don’t see it.

Ben:Yeah, if you changed the x-axis and didn’t know it was years of practice, you might think these are different people altogether. People with more than 20 years in practice strongly disagree with certain degrees of uncertainty, a response you almost never see in fellows or early-career doctors. Maybe that brings hope? You feel anxious because you’re early in your career. If you stick with it a little longer, hopefully experience will help.

Daphna:Maybe. If you don’t get burnt out...we need you all to stick around.

Ben:If you don’t burn out and quit medicine altogether in the meantime.

Daphna:Thank you for sharing. This paper did bring me some solace.

I also have two papers quickly that I’ll try to work through. I don’t do anything quickly, but I’ll try. This one was interesting, it was in the Journal of Pediatric Gastroenterology entitled “Hydrostatic Low-Volume Enemas in Infants with Birth Weight ≤ 1,000 grams or Gestational Age ≤ 28 weeks: A Controlled Interventional Study.” Lead author Tabea Stock, senior author Andreas Jenke, coming from Germany. It’s a small study: 42 infants in the control group and 74 in the intervention group. The intervention looked at standardizing saline enemas. I thought this was interesting because in the States and globally, some people use enemas, some use glycerin, some use nothing. We don’t really know the best way to evacuate stool in babies. In this unit, people were giving a lot of enemas, but with no standardization.

It was one group from January 2019 to June 2020. Infants had to be born at the hospital, stay longer than 24 hours, and have delayed meconium passage, meaning absence of meconium for more than 24 hours during the time when the infant only passed meconium-type stools (not transitional stools yet). They were eligible for enema if no bowel movement for over 24 hours. Enemas had to happen at least once in the first 20 days of life.

In the control group, physicians performed enemas manually, using normal saline of varying volume. Catheter was inserted “as deeply as possible,” and timing and frequency were based on physician assessment. The intervention group used a new, minimally invasive protocol with a 6 Fr urinary catheter, a three-way stopcock, and syringe without its plunger to vent. Then, 10 ml/kg saline poured into the syringe with the plunger, fixed at 15 cm to achieve a pressure of 15 cm water. The catheter tip was lubricated and inserted rectally 2.5 cm, the stopcock was opened, and the plunger was removed, allowing the solution to run through.

Enemas were only applied if there was no bowel passage for more than 24 hours or if the last enema was more than 24 hours ago. Feeding regimen was consistent with most practices, so I won’t get into that.

Primary outcome was protocol safety assessed by GI complications (e.g., NEC, SIP, meconium plug syndrome). Secondary outcome was effectiveness of evacuating stools and nutrition parameters at six time points (days 2, 4, 6, 10, 15, 20). They documented first and last meconium, number of stools, time to full enteral feeds (120 ml/kg/day), need for parenteral nutrition, and last day of TPN use.

They had 42 infants in control, 74 in intervention, but 30% protocol violations in intervention (defined as 1 or more wrongly performed enemas). That’s just something to note, they even have pictures showing when it’s done wrong. Gestational age was similar in both groups. Birth weight was significantly lower in the intervention group compared to the control group (870g vs. 957g). Some characteristics differed: preterm premature rupture of membranes was more common in the control group, HELLP syndrome more common in intervention, and “other reasons” also more common in intervention.

They looked at respiratory support. Median oxygen saturation on day 4 was 96% in control, 93.9% in intervention (statistically different) but no other time points. On day 20, 37.5% of control infants breathed unsupported versus only 18% in intervention (this was significant). More infants in the intervention group needed antibiotics on days 4, 6, and 10. A significantly higher weight gain was seen in week two in the intervention group.

In control, 90% had at least one enema vs. 97% in intervention. Fewer enemas per child were received in the intervention (4 vs. 8). First meconium was comparable, however last meconium occurred later in intervention group (7 vs. 6 days). Fewer stools passed in the first 20 days of life in the intervention group (61 vs. 72). Feeding tolerance was better in intervention group; more infants on full enteral nutrition on day 10 (31% vs. 14%). TPN intake was higher in intervention in the first four time points. Percent needing TPN and IV fluids and duration of parenteral nutrition were similar.

No significant differences in primary outcome. NEC occurred in 4% intervention vs. 9.5% control (not significant). SIP 2.7% vs. 7.1% (not significant). Meconium plug syndrome 6.8% vs. 2.4% (not significant). There was a trend (p=0.088) for reduced intestinal morbidity in intervention (6.8% needing surgery vs. 16% control). Mortality rates were similar. Combined mortality and morbidity were lower in intervention (6.8% vs. 19%), but not statistically significant.

I thought it was interesting to have a standardized protocol, though I’m not sure it changes practice.

Ben:I don’t know if I have much use for the intervention as is. Many of us use glycerin suppositories or enemas to stimulate stool passage. I would have liked that to be the control group, instead of saline enemas as control versus new method. The cohort is very small. Rates aren’t that different. They also have a pretty high baseline NEC rate at 9.5%, which may be higher than many are used to. But we know preemies have delayed or abnormal gut motility. And the question is, should you try to evacuate the intestine more thoroughly? Would that improve outcomes? I don’t know.

Daphna:Feels like it, but we don’t have that answer. Does it even help? All these things we do—do they?

Ben:Yeah, and is stool stasis a recipe for bacterial translocation, i.e., NEC? Good question to study.

Daphna:Are you gonna let me do my last one? I’m gonna be so fast. Okay, this one is “The New Normal: Parental Use of Online Health Communities in the NICU.” We’ll hustle through this. It was discussed in a NeoNews episode recently. Lead author Shannon Adams, senior author Beatrice Lechner, coming from Brown University. They wanted to identify if and why NICU families use online health communities and how that use impacts relationships between parents and their child’s medical teams.

They did a two-part study: first, they distributed surveys to NICU parents selected through an internet search identifying NICU support or advocacy groups (groups were defined as “active” if there was at least one post in the last three months) that were targeting NICU families. They also distributed surveys via the NICU Families Reddit stream with permission from their administrators over three months. The second part involved collection and evaluation of NICU-centered content posted on that Reddit group. The NICU sub-Reddit was chosen because it has 4400+ subscribers (but thousands more readers), and it’s publicly accessible (unlike Facebook), allowing confidentiality since users don’t have to use real names.

In total, they had 250 people participating in the survey, but had to remove a few, leaving 242 surveys. 60% of the people who were already part of these groups answered “yes” when asked if they used specific social media sites or online health communities for NICU parents and families to discuss their child in any way during their NICU stay. They were then given a list of 10 reasons why they may have used those communities. The most commonly selected responses were: 78% wanted to make contact with other NICU families with similar problems, 67% wanted to reduce anxiety regarding information given to them by their medical team, 63% wanted to share their success with other NICU families with similar experiences, 53% wanted to vent frustrations to other NICU families with similar experiences, and 44% wanted to know more about the information provided by their medical team.

61% noted an overall positive effect on communication with the team. 56% noted an overall positive effect on trust in the team. 60% noted an overall positive effect on confidence in their team. So they were communicating online, but how did it affect their relationship with their home team? Nearly 60% documented that the use of these online health communities had an overall positive effect on the general quality of their relationship with their child's NICU team. Very few reported negative effects. 2.8% reported a negative effect on communication, 4.3% reported a negative effect on trust, 2.8% reported a negative effect on confidence.

When they looked at Reddit posts, they analyzed 176 posts extracted over two one-month periods, each with varying levels of engagement. The posts were divided into five subcategories: (1) Requesting or sharing medical advice, (2) Venting frustrations regarding the medical care team or setting, (3) sharing positive news or accomplishments, (4) seeking general support or advice, and (5) other.

In total, 30% of posts were seeking general support or advice, 27% were sharing positive news or accomplishments, and 26% were requesting or sharing medical advice. Interestingly, those asking for medical advice were willing to share detailed personal health information about their child, including gestational age, birth weight, and diagnoses, to get targeted advice.

They didn’t dive into all categories in detail but wrapped up the discussion nicely. The most common reasons for being on these communities were a desire to connect with other families or a need for additional resources. Again, 78% wanted to connect with other families, 63% wanted to share successes, and 53% wanted to vent frustrations. However, the frustrations weren’t always with the NICU team—they were often just about the NICU admission itself. Only 11% said they had questions unanswered by their medical team, and only 6% said they did not understand information provided by their medical team. So it seems like the medical team is doing a good job giving information and that families understand it, but families want more. 66% wanted to reduce anxiety regarding information given by their team. 44% wanted to know more about the information provided by their medical team.

The common themes in posts were (1) Vulnerability — 30% of posts included photo content, often of the baby themselves and (2) consistent engagement with other members of the subreddit. Across 176 posts, there were 2,154 comments, averaging 12 comments per post. Parents shared posts, asked for advice, and revealed frustrations, which were consistently met with kind words, encouragement, and well wishes.

Ben:I think those categories were interesting because if you classify posts that way, it makes sense why social media or Reddit platforms can feel so toxic. If you're a parent in the NICU and the posts you see are mostly people venting frustrations, that’s tough. Sharing positive news might not feel great either if you’re struggling and see other babies doing well. Posts sharing or requesting medical advice or seeking support aren’t always reassuring either. So it’s no surprise that some responses had a neutral tone about the themes. I don’t think we need to reform how social media is used, but I do think it’s important to understand that local NICU support groups — like the one you’re leading that was recently featured in the news down here in Florida — are probably the right way to do it.

Daphna:Yeah, that’s an interesting perspective. I appreciate it, thank you. Alright, buddy, we made it!

Ben:Yeah, we made it. Thanks everyone for listening, and we’ll see you this week with more Incubator content.