top of page
Horiz_red_podcast.png

#308 - 📑 Journal Club - The Complete Episode from May 11th 2025

Updated: May 16




Hello friends 👋

In this special anniversary edition of the Journal Club, Ben and Daphna celebrate four years of The Incubator Podcast while diving into a compelling lineup of neonatal studies. The episode kicks off with a review of a phase 2 multicenter trial on the safety of furosemide in preterm infants at risk for BPD. Despite widespread Lasix use in NICUs, data on dosing and safety have been lacking—this study finds no significant increase in serious adverse events but emphasizes the need for larger trials to better define its role. The team then explores a study from India comparing 7- vs. 14-day antibiotic courses in culture-proven neonatal sepsis, showing that shorter courses may be safe and effective in select populations.


Additional discussions include a randomized trial from Australia evaluating “sigh breaths” during high-frequency oscillatory ventilation and their effects on lung volume and oxygenation, a study examining how kangaroo mother care boosts breast milk intake, and a large Japanese cohort study detailing neurodevelopmental outcomes of infants born at 22–31 weeks. Finally, a meta-analysis on prenatal cannabis exposure underscores risks of low birth weight and preterm birth. With depth, humor, and clarity, Ben and Daphna guide listeners through evidence that shapes neonatal care.


----























We’re celebrating four incredible years of The Incubator podcast! As a thank-you to our amazing global community of listeners, we’re hosting a special giveaway packed with awesome prizes. Entering is easy — just take a super short 2-minute survey about your experience with the podcast. By filling it out, you’ll be automatically entered to win incredible prizes including a MacBook, Bluetooth headphones, Bluetooth speakers, Incubator merch, books, tickets to the Delphi Neonatal Innovation Conference, and so much more. Your feedback helps us grow, improve, and continue to serve the neonatal community. Thank you for being part of this journey — and good luck! Click here to enter


----


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


Kono Y, Kusuda S, Nishida T, Mori R, Toyoshima K, Sasaki H, Yonemoto N, Uchiyama A, Fujimura M; INTACT Study Group.J Perinatol. 2025 Mar 26. doi: 10.1038/s41372-025-02268-3. Online ahead of print.PMID: 40140596


Monnelly V, Nakwa F, Josephsen JB, Schmölzer GM, SolevÄg AL, Rabi Y, Wyckoff MH, Weiner GM, Liley HG; International Liaison Committee on Resuscitation Neonatal Life Support Task Force.Arch Dis Child Fetal Neonatal Ed. 2025 Apr 29:fetalneonatal-2025-328577. doi: 10.1136/archdischild-2025-328577. Online ahead of print.PMID: 40306761


Martinez TA, Remmer E, Mardakis S, Leone M, Boyer J, Lv S, Beltempo M, Sant'Anna G, Shalish W.J Perinatol. 2025 Apr 12. doi: 10.1038/s41372-025-02291-4. Online ahead of print.PMID: 40221609


Lo JO, Ayers CK, Yeddala S, Shaw B, Robalino S, Ward R, Kansagara D.JAMA Pediatr. 2025 May 5:e250689. doi: 10.1001/jamapediatrics.2025.0689. Online ahead of print.PMID: 40323610


Sinha B, Mazumder S, Thakur A, Devi S, More D, Ashorn P, Sommerfelt H, Kurpad A, Bhandari N.Am J Clin Nutr. 2025 May;121(5):1109-1116. doi: 10.1016/j.ajcnut.2025.02.006. Epub 2025 Feb 11.PMID: 39947481 Free article. Clinical Trial.


Hough JL, Jardine L, Hough MJ, Steele M, Greisen G, Heiring C.Arch Dis Child Fetal Neonatal Ed. 2025 Apr 17;110(3):297-302. doi: 10.1136/archdischild-2024-327445.PMID: 39406472 Clinical Trial.


Dutta S, Nangia S, Jajoo M, Sundaram M, Kumar M, Shivanna N, Gathwala G, Nesargi S, Jain S, Kumar P, Saili A, Karthik A, Tripathi S, Bandiya P, Dalal P, Ray P, Randhawa VS, Saigal K, Radhakrishnan D, Venkatesh V, Jagannatha B, Sharma M, Nagaraj S, Malik M, Dogra S, Mittal S, Saini A, Makkar N, Dhir M, Chandramohan A, Pragati RA, Srivastava T, Mukundan L, Benakappa N, Shukla A, Rasaily R.Arch Dis Child Fetal Neonatal Ed. 2025 Apr 25:fetalneonatal-2024-328232. doi: 10.1136/archdischild-2024-328232. Online ahead of print.PMID: 40280737


Greenberg RG, Lang J, Smith PB, Shekhawat P, Courtney SE, Hudak ML, Moya F, Iyengar A, Eldemerdash A, Bloom B, Go M, Hanna M, Rhein L, Aliaga S, Lewis T, Febre A, Kiefer AS, Bhatt-Mehta V, Khoury JA, Selewski D, Anand R, Martz K, Payne EH, Zimmerman KO, Benjamin DK Jr, Laughon M; Best Pharmaceuticals for Children Act – Pediatric Trials Network Steering Committee.J Pediatr. 2025 Apr 28:114629. doi: 10.1016/j.jpeds.2025.114629. Online ahead of print.PMID: 40306549


----


Watch this week's Journal Club on YouTube




----


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


Ben Courchia, MD

Hello everybody, welcome back to the Incubator podcast. We are back for an episode of journal club. Daphna, how are you?


Daphna Yasova Barbeau, MD

I'm doing well buddy. I've got everything pulled up here. Trying to keep the computer quiet so we can have a nice seamless episode.


Ben Courchia, MD

This episode is going to be released in the podcast, obviously, but also on YouTube. We're trying to do a lot more video stuff for you guys. That being said, we're at this point of the year where we're celebrating the anniversary of the podcast: May 4th, 2021 was the first episode. It's now been four years that we've been doing this. It's truly been four years. We haven't missed much.


Daphna Yasova Barbeau, MD

No, hardly a week, I think, in four years.


Ben Courchia, MD

We will roll out this week on the website our annual giveaway to celebrate the birthday of the podcast. We've talked about it on YouTube, and we will repeat it here. We're going to do the same thing we did last year. It's a very short survey. We take that information. It helps us make the Incubator better. If you submit the survey, you automatically enter to win one of our many prizes. We have an Apple computer this year. That's very exciting. We also have some swag from our shop. We also have tickets to Delphi. We have books. I don't know if everybody's going to get a prize, but at least you should feel like you have a high chance of walking away with something.


Daphna Yasova Barbeau, MD

We like to give stuff away!


Ben Courchia, MD

It will be one of the featured pages on our new website, and then you can go fill out the survey. It'll literally take you five minutes, and then you're entered. You don't need to do anything. We used to do a lot of these tweet this, repost that, hold one hand up and one foot down



Daphna Yasova Barbeau, MD

Do a handstand.


Ben Courchia, MD

No. Go to that survey. You're done.


Daphna Yasova Barbeau, MD

We really appreciate when people fill out the survey because we have always made changes, we read every single comment. We make changes. We've always made incremental change based on the feedback from the community. As we have our fourth birthday, it's really in part to everybody who tunes in and listens and engages with us. We're so grateful for people who have been along for the ride, who are joining us now, or sharing with a colleague.


Ben Courchia, MD

Whenever we hear that this works for you all (not that we need to hear that it works), but when we hear that what we're doing actually makes a difference in your day to day, it gives everything we're doing meaning the meaning that we're hoping for.

That being said, Journal Club doesn't wait for any man or woman.


Daphna Yasova Barbeau, MD

I know you're very excited about this group of articles.


Ben Courchia, MD

Yeah. I'm going to start with something that is probably on everybody's mind all the time this is a paper that is published in the Journal of Pediatrics and it's called, “Furosemide Safety in Preterm Infants at Risk for Bronchopulmonary Dysplasia.” First author is Rachel Greenberg. It's a very interesting paper. Taking a look at furosemide (Lasix), we know that Lasix obviously is used in so many units. Half of all preterm infants under 1,000 grams in the NICU get Lasix. It's that stat that keeps coming back with every furosemide paper, but it still every time blows me away that we still don't have a good idea on how and when to use it. It's interesting how our field is! The idea in this paper is simple. We try to treat pulmonary edema, we try to improve gas exchange, but can we reduce the incidence of BPD? Until now, we've only really had small trials, maybe database review, retrospective reports; there's not a lot of pharmacokinetic data available out there and the results are what we are encountering – massive variability.

This particular study is important because it's going to be the first multi-center randomized placebo-controlled dose-escalating trial of furosemide in this particular population of preterm infants. I was very interested in reading this paper. It's a phase two trial that was conducted between 2015 and 2018 across 17 NICUs in the US. They included babies based on certain eligibility criteria. They had to be less than 29 weeks of gestation at birth. They had to be 7 to 28 days postnatally at randomization, and they had to be on some form of respiratory support (CPAP, high flow, or mechanical ventilation). Their exclusion criteria included diuretic exposure, electrolyte or renal abnormalities, hemodynamically significant PDA, meconium aspiration, or known diuretic allergies (which I've never thought of before).

The intervention is an interesting one. They randomized kids in a 3:1 ratio of furosemide to placebo, and they divided their furosemide group into three groups. So bear with me here. They had the first cohort, which was a low dose. They give them 1 mg/kg IV every 24 hours. They did also offer enteral, and the conversion was double. They had this pharmacokinetic assessment that determined that if you were using 1 mg/kg IV, based on the absorption, it would need to be 2 mg/kg enteral. Every dose will be multiplied by 2. Then they had a cohort number two, which was a medium dose that was 1 mg/kg IV given every six hours. Cohort number three was the planned high dose cohort. “Planned” is the keyword here, because they were supposed to give 2 mg/kg IV Q6 hours. But they never really initiated that, because there were some safety concerns. We'll get into that. The infants in the study all received the drug for up to 28 days; whether it was going to be given IV or oral was left at the discretion of the team. They monitored the labs for these babies pretty closely, electrolyzed sodium, potassium, bicarb, BUN, creatinine, and phosphorus and they checked at least twice weekly following dose escalation
I say BUN. Do you say “bun?” I've heard people say both and the first time I heard “bun,” I was like, what is that?!

It's a phase two trial, so they measured safety in the form of adverse events or serious adverse events. But they also had some key outcomes that included death before 36 weeks, moderate to severe BPD or death, electrolyte abnormalities, hearing loss, and then renal complications like nephrocalcinosis or nephrolithiasis.

Let's get into some of the results. They ended up having about 82 infants randomized. 62 went into the furosemide groups and 20 went into the placebo group. 80 completed the study. The median gestational age was 26 weeks. The median birth weight was 798 grams, and the median postnatal age at randomization was 21 days. 93% in the furosemide group reach their max dose. I'm not going to get too much into that, but for the kids in the higher dose, they did a dose escalation process where they started them on lower doses and then eventually started escalating the dose, checking electrolytes, making sure that this was well tolerated. It's important to know that 93% actually reached their max dose. There was also some open labeled diuretic use in the study, like 24%. Take that as you wish.

In terms of adverse events, surprisingly, even in the placebo group there was a lot of reported adverse events. [ A total of] 293 adverse events in 74/80 (93%) infants were noted, including 223 adverse events [among 56/61 (92%) infants who received furosemide and 70 adverse events among 18/19 (95%) infants who received placebo]. So there were adverse events in both groups, but most of them were mild or moderate. There were [a total of] 14 serious adverse events in 20% of infants in the furosemide group, and only 11% in the placebo group. Now of these 14 severe adverse events in the furosemide group, two were deemed by the site investigators as at least possibly related to the study drug. Definitely something that to me doesn't seem particularly striking and speaks to the safety furosemide; we'll talk a little bit more about that.

When it comes to some of the secondary endpoints, let's talk about mortality and BPD. On adjusted analysis, they found no significant differences in the combined outcome of BPD or death between the groups, with an odds ratio of 0.97 for cohort one compared to the placebo and an odds ratio of 2.59 for cohort two versus placebo. In terms of serum electrolytes, they found one or more serum electrolyte adverse events in 83% of the 80 infants (84% of the cohort receiving low dose of Lasix, 93% of the cohort receiving moderate dose of Lasix). On adjusted analysis, the odds ratio for having any serum electrolyte abnormality was 4.46 in cohort one compared to placebo, and 7.89 for cohort two (medium dose versus placebo).

Another interesting outcome is hearing loss. They found that hearing loss prior to discharge was seen in 14% of the furosemide-treated infants and in 22% of the placebo-treated infants. As you can suspect on the adjusted analysis, they found no difference in furosemide potentially increasing the risk of hearing loss, which was surprising because we know that ototoxicity is something that we always learn about furosemide. Could it not have an impact in babies this small? Maybe. The authors do talk about that in the discussion.

In terms of nephrocalcinosis and nephrolithiasis, nephrocalcinosis was diagnosed on renal ultrasound, and it was seen in 23% of the furosemide-treated group compared to 29% in the placebo. Again, not something that was found to be statistically significant. The nephrolithiasis was also not significantly different, but the numbers were very, very small in that case. They're having some issues, reporting on that and making it more of a generalizable result.

In terms of BPD or death risk, which is also an innovative thing that this study did, where they didn't just look at the outcome of BPD, as we mentioned earlier, but they also looked at the potential trajectory. Using the BPD calculator available on the Neonatal Research Network website (on Neonatal Research Network website, you can go under “tools” to get the outcomes estimator for your prenatal consult, but you can also get the BPD risk calculator). The BPD or death risk (because death is one of the outcomes that is measured in the BPD calculator) decreased from baseline to week four, regardless of the study drug. On adjusted linear regression analysis, the treatment estimate for cohort one versus placebo was (-)2.7, and for cohort two it was (-)1.34. The overall treatment effect was not found to be significant with a p-value of about 0.6.

Now, in terms of the third cohort with the very high dose, the trial was terminated after an unmasked review of cohort one and two interim safety and efficacy data. The FDA concluded that the risk-benefit assessment for the third high dose group was not favorable, and directed the study team to remove the high dose cohort, which was what they did. In summary, this trial of preemies shows that Lasix (furosemide) administration at doses of one through four mg/kg/day for 28 days was not associated with an increased incidence of adverse events. It was not associated with an increase in hearing loss, nephrocalcinosis, or nephrolithiasis, and future larger trials are needed to better define the optimal use of Lasix in preterm infants with developing BPD. Now these trials will likely need to include management of electrolyte as part of the trial design with amelioration of the electrolyte abnormalities. These trials may be able to safely administer higher doses of furosemide to determine the risk and benefits of prevention of BPD.

I think that this is obviously something that in retrospect feels a little bit obvious that yeah, like Lasix is going to mess up your electrolytes, but obviously that cannot be sufficient to exit the trial because then you're going to have issues again trying to test these much higher doses. I thought this was interesting. I don't think it changes very much. I'm not a big advocate for Lasix in general, but it's data that needs to be reviewed if you're a neonatologist. It is after all a phase two multi-center trial randomized trial. It's important data and we should be familiar with that.


Daphna Yasova Barbeau, MD

There was a big pendulum swing on diuretics in general, right? Revisiting this is interesting. It's a lot of Lasix. So here we are, like, should I do a three-day course, a five-day course, can the baby get just one dose? And they're doing 28 days! It was interesting to say the least. I was also quite impressed by how much nephrocalcinosis there was in the placebo group.


Ben Courchia, MD

That's always the big fear to me –we don't know the effectiveness of Lasix, but what about the side effects? But to me, to see that hearing loss was not as prevalent as I would have expected, neither are kidney stones, I think that's quite good. We've spoken to Nick Bamat from CHOP at Hot Topics about [Lasix]. He had published some work using the Pediatrix database on Lasix showing that, number one, Lasix remains one of the most used medications in the NICU. We should have a conversation about Lasix. But again, advocating for what you were saying, that it probably has a place, but it doesn't have a place for an indefinite type of treatment where kids can be on it for weeks and weeks and weeks with no end point in sight.

Daphna Yasova Barbeau, MD

Yeah, very interesting. All right, I have one for you. I have some international papers this week. You put this one into the folder, “Seven-day versus 14-day antibiotic course for culture-proven neonatal sepsis: a multicenter randomized non-inferiority trial in a low and middle-income country.” This is coming from India. The study was asking, is seven days non-inferior to a 14-day course, even then the face of culture proven infection? This was a multi-center randomized non-inferiority trial, one-to-one allocation ratio, and the outcome assessment was felt to be blinded. This is part of an ongoing study; this is like a sub-study.

They enrolled neonates that were zero to 28 days old with a birth weight of greater than or equal to a thousand grams (not the smallest group of babies, but still quite small) and suspected sepsis. Antibiotics were initiated by a neonatologist after sending a blood culture. This is a group of babies that were screened. But inclusion criteria required a positive bacterial blood culture, receipt of at least seven days of “appropriate antibiotics” that the pathogen was unambiguously sensitive to, and clinical remission of signs of sepsis by day five of antibiotics. On day seven, the baby had to be improved and well-appearing. And on day seven, they were randomly assigned to either receive the seven days and complete antibiotics, which was the intervention to be done, or receive seven more days for a total of 14 days, which was the comparison arm for antibiotics.

They had some exclusion criteria: the growth of contaminants (they excluded staph aureus, interestingly, or fungi), if there was ambiguous in vivo antibiotic sensitivity, or infections that would have felt to necessitate more than 14 days of antibiotics like meningitis or life-threatening malformations. A lumbar puncture was performed on all participants. CONS (coagulase negative staph) was considered a contaminant if it grew after 48 hours of incubation, or if repeat cultures were sterile despite resistance to the empiric antibiotics chosen. They also stratified the randomization by center and birth weight category: 1000-1500 grams, 1501-2000 grams, and then greater than 2000 grams. They felt that weight would be potentially an important determinant of post-immune response. They enrolled the babies and they monitored them weekly for at least 35 days post-randomization.

The outcome variable they were looking for primarily was this composite of a definitive or probable relapse within 21 days after antibiotic completion. Did we treat it, or did we only partially treat it and it came back? A definitive relapse was an episode of sepsis caused by the same species as the original infection with an identical antibiogram. A probable relapse was adjudicated based on clinical signs and laboratory parameters with a sterile blood culture. They had a definite relapse or if the baby had these new signs and symptoms, but they couldn't prove the same species after treatment.

They recruited participants from February 2019 to December 2022. They actually performed an interim analysis after 261 patients were enrolled (126 in the seven day group and 135 in the 14 day group) and had completed their follow-up. After the interim analysis, the data safety monitoring board recommended trial termination, as they felt non-inferiority was demonstrated using two-sided 99.5% confidence intervals.

About the study, the baseline characteristics were pretty similar between the two groups. In the 14-day arm, 10 participants clinically deteriorated during the eighth to 14th day of antibiotics. They were on the extended course of antibiotics, including two with new culture positive sepsis; one died and one was transferred. Six participants were excluded from the follow-up analysis: one from the seven-day arm because that baby was enrolled despite probable meningitis, and five from the 14-day arm, because four were lost to follow-up and one had an incomplete antibiotic course.

In the 14-day group, one infant required ventilation and inotropic support, one required a packed red blood cell transfusion. None in the seven-day group required ventilation, inotropes, or blood products after randomization. Two participants in the seven-day arm and six in the 14-day arm experienced the primary outcome with a risk difference of (-)3%. The confidence intervals cross zero. As the upper bound of the confidence interval was well beyond the non-inferiority margin, there was a recommendation to stop the trial.

In the modified intention-to-treat analysis, they included three additional participants whose outcomes were known, one lost to follow-up but later identified, one whose 14-day course was incomplete, and the one enrolled despite meningitis. Three others were excluded from that analysis, because then they were lost to follow-up and could not be tracked. But the risk differences for the primary and secondary outcomes supported the non-inferiority of the seven-day antibiotic course.

They did a sensitivity analysis for many outcomes. I will list them for you: definite or probable relapse by 21 days, mortality or definite or probable relapse or secondary sepsis by 21 days, definite or probable relapse by 28 days, mortality or definite or probable relapse or secondary sepsis by 28 days, definite or probable relapse by 28 days. They use those both post-completion and post-randomization, so they had duplicate sets here. All of those sensitivity analyses favored the non-inferiority of the seven-day course. The median durations of secondary outcomes were significantly lower in the seven-day arm, and adverse events did not differ significantly between the two groups.

I think this is interesting. We're always talking about antibiotic exposure I think that's true in a lot of facets of medicine. We don't know how much antibiotic is enough and we are reducing antibiotic courses for lots of things. It would be nice if we could use shorter courses in our small babies. I'm not saying people should change what they're doing. I'm just saying, I'm glad people are looking into it, because hopefully we'll get to a place where we can use shorter courses.


Ben Courchia, MD

I agree, and I think that it's going to come from low and middle income countries, where resources are definitely thought of a little bit more carefully. Maybe they will be able to bring about the reassurance that a shorter course, even in culture-proven sepsis can actually be valuable.

I have definitely two papers that I want to bring. Maybe I'll leave this fun one for last. The other one that I wanted to review is a paper that was published in the Archives of Disease and Childhood called “Intermittent sigh breaths during high-frequency oscillatory ventilation in preterm infants: a randomized crossover study.” I feel like it's an ongoing question. I don't know where some of our listeners are practicing, but sigh breaths on high frequency is always a point of discussion. Should we go up? Should we go down? Should we even have it in the first place?


Daphna Yasova Barbeau, MD

Even in our own unit.


Ben Courchia, MD

Oh my God, yes. We know that the reason it's a topic of conversation is because ventilator-induced lung injury is always a concern. One of the goals with high frequency is to maintain adequate and expiratory lung volume, while minimizing barotrauma and volutrauma. The sigh breaths are widespread, and it's been used obviously for patients who are on high frequency jet ventilation, but there's a growing interest in combining brief conventional sigh breaths for high frequency oscillatory ventilation to facilitate lung recruitment without interrupting the high frequency pattern. This paper is actually here to test this idea in the form of a randomized clinical trial.

This study aimed to answer the question using electrical impedance tomography or EIT. I have started to learn more about it over the past 12 months. It's a fascinating tool where you have a non-invasive bedside tool that can track changes in regional and global lung volume, including end-expiratory lung volume by measuring impedance across electrodes placed on the chest. The kids wear a belt around their chest, and you get these diagrams of long volumes and aeration. It's quite interesting.

This was a randomized double crossover trial conducted at the Mater Mother's Hospital in Brisbane, Australia. It's a short, small study; 16 preterm infants, all less than 36 weeks gestational age were enrolled. They were all on high frequency oscillator, and they had to have their FiO2 between 25 and 70% to maintain sats above 90. Major congenital lung or heart anomaly were considered exclusion criteria. They randomized these infants into two groups, some of them receiving high frequency oscillation with a sigh breath, some without. The sigh breaths were delivered three times per minute, so the rate was three. Each sigh used an inspiratory time of one second and a PIP of about 30 centimeters of water. The sigh frequency was chosen to limit interference with CO2 clearance, maintaining the gas exchange strategy. The primary outcomes were change in end-expiratory long volume and ventilation distribution measured through the GI index. Secondary outcomes looked at the sats, the heart rate, the blood pressure, and other ventilatory parameters.

The results are very interesting. The infants in the study had a median gestational age of about 25.5 weeks and a median birth weight of about 700 grams. Truly like small babies, which was helpful. What the team found was that the addition of sigh breaths significantly increased global end-expiratory lung volume. This increase was especially notable in the posterior or the dependent lung regions. That was found to be statistically significant and also in the left lung. Over time, the increase in global end-expiratory lung volume, right and left, and the GI index improvement were all statistically significant. There was a trend toward increased impedance amplitude, suggesting possible tidal volume benefits as well.

From a physiologic standpoint, the oxygen saturation was significantly improved on the group that received sigh breaths. The mean airway pressure was not significantly altered, if anything, MAP set points were actually lower in the group that received sigh breaths. The relationship between sats and FiO2 showed no statistically significant improvement, but the overall direction seemed to be positive. There were no other significant changes in hemodynamic or ventilatory parameters between the two strategies.

In conclusion, the authors are saying that sigh breaths during high frequency oscillation ventilation improves end-expiratory lung volume, ventilation distribution, and short-term oxygenation, and these findings are physiologically sound and clinically promising. However, the study is very short in terms of the number of babies that are enrolled. It's not looking at long-term outcomes and it's definitely not powered for neurodevelopmental outcomes. We don't have the safety data on extended side breath usage and its effect on outcomes like BPD and ventilator days, so that's why they're recommending longer and larger trials. But I thought this was interesting. I saw the title and [thought that I could bring] that back to the table.

Daphna Yasova Barbeau, MD

Yeah. You could see how potentially over time that may be helpful, but yeah, we'd have to balance it with those adverse events and things like that.

Okay. I actually have another study from India, which I'm just going to, just do briefly, because I think it's a value, but their practice is very different than ours. So this is in the American Journal of Clinical Nutrition. It's called “Effect of Kangaroo Mother Care in Low Birth Weight Infants on Human Milk Intake, a Randomized Control Trial.” The study participants were part of another primary trial, but this is another sub-study to look at eligible babies in terms of kangaroo care. They took newborns weighing 1500-2250 grams and their mothers and screen them from enrollment within 72 hours of birth in the first trial. In the sub study, they were able to consent moms and their infants aged between days three and seven after birth. They have a group early on from 2015 to 2018, and then they even enrolled more babies from 2017 onward.

What they did is they had an intervention group and they had a control group. The Kangaroo Mother Care Intervention Group was comprised of promotion and support of continuous skin-to-skin care and exclusive breastfeeding. The intervention delivery team would visit the infant-mother dyads allocated to the intervention arm; most of these visits were occurring in their homes, even in this cohort of babies, 1500 grams to 2250 grams, so they were going and supporting kangaroo mother care with the dyad on days one, two, three, five, seven, 10, 14, 21, and 28 after birth, to really just help them manage with this continuous skin-to-skin care. They were counseled to practice skin-to-skin care for as long as possible during the day and the night, and they trained the other family members to help assist this mother.

They wanted to really look at the mean daily infant human milk intake, and they wanted to look at some other composition of breast milk markers or nutritional biomarkers. They used deuterium dilution dose to mother technique. The mom was given this measurable drink of deuterium at their homes, then they were able to pull it from the saliva in the moms and the babies. That's how they measured how much milk transferred. They were using this mean daily water transfer from mother to infant during a 14 day period.

 

They had 550 eligible infants and their mothers enrolled and they wanted to look at milk assessment. Among the 550 participants, infant human milk intake assessment was completed in 85% in the control arm and 86% in the intervention arm. The median age of enrollment was day five in both arms. The median infant weight at enrollment was 2.1 +/- 0.2 kilos. Baseline characteristics were similar across the arms for the maternal factors. Place of birth, like home births, was thought to potentially be a confounding factor, and it was unequally distributed: 18% in the control arm and 13% in the intervention arm delivered at home. In both study arms, 64% of the included infants were born preterm. There were no significant differences in the baseline characteristics of the infants, when those data was available.

In the control arm, so not the intervention arm asked to do skin-to-skin care, 8% of mothers reported practicing skin-to-skin care and the mean duration of skin-to-skin practice was two +/- seven days with 0.5, so 30 minutes +/- two hours per day. In the intervention arm, all the mothers reported practicing the use of skin-to-skin care. The mean duration of skin-to-skin practice was 28 +/- two days with a mean duration of 12 +/- four hours a day. The mean human milk intake in infants over the 14 day period from enrollment was 331 +/- 144 in the control arm, and 368 +/- 135 grams per day in the intervention arm.

So compared with the control arm, the mean difference in infant human milk intake in the Kangaroo Mother Care arm participants, adjusted for place of delivery, was 37 grams per day or 12 grams per kilo per day. On the basis of multiple analysis, the mean daily intake of human milk was 331 grams per day in the control arm, 370 grams per day in the intervention arm, yielding a mean difference of 38 grams. In the subgroup of infants born preterm, the adjusted mean difference in the human milk intake was 45 grams per day, compared with 20 grams per day in full term infants. So, the effect was greater in those babies born preterm. Among SGA infants, the adjusted mean difference in human milk intake was 40 grams per day, whereas in appropriate for gestational age infants, it was 31 grams per day. So again, a bigger effect in the SGA infants. The mean concentration of carbohydrates in human milk for mothers in the control arm was 50 +/- six. In the intervention arm, it was 51.2 +/- six. The mean difference was 1.2. The mean concentration of total protein in human milk was 16.3 +/- four then the intervention arm was 15 +/- two. The mean difference was minus 1.3. The concentration of fat was similar in the control arm and the intervention arm. They did not find any differences in the mean concentration of the human milk amino acids, IgA concentrations, and lactoferrin in the study arm.

So I thought this was interesting because, well, it's a good reminder that in many places of the world, moms are doing 12 hours of skin-to-skin care a day. This could potentially improve milk output and it had a bigger effect for preterm babies and SGA babies.

Ben Courchia, MD

It's always these types of studies where you're mostly not surprised, but you're like yeah, sign me up. We all have constraints when it comes to resources, but it's interesting also to see how other countries leverage these very cheap and low-tech tools to help you achieve outcomes that go beyond that.

All right. We don't have a lot of time left. I'm going to do the paper that I've enjoyed reading the most this week. This is a paper that I found in the Journal of Perinatology and it's called “Interprofessional perspective on non-invasive respiratory support practices in extremely preterm infants: a Canadian survey.” I'm on good terms with Dr. Marc Beltempo and Dr. Guilherme Sant'Anna, so I was curious to read this paper just because they wrote it, but also because of the title. We all know that non-invasive respiratory support is critical for us in the NICU, but while it helps us avoid intubation and associated complications, its application is far from straightforward and it demands bedside attention. You have to maintain it. You have to monitor. You have to troubleshoot. It's a lot of work. Failure of noninvasive respiratory support is not a small deal. You have both short and long-term morbidities that are associated. The authors were asking the following question: can we better understand the barriers and the variability that is associated in practices related to noninvasive ventilation?

They did so by listening to the frontline workers, physicians, nurses, and respiratory therapists. They sent this survey nationally in Canada. They identified 32 NICU, so they sent it to their 32 NICU directors. They sent it to 13 fellowship directors, 32 nursing managers, and 32 respiratory therapist-educators, who then circulated that survey with their teams. It had 19 items in three categories: (1) the perspective of people on non-invasive respiratory support devices and management strategies, (2) the monitoring and management, and (3) the failure and improvement strategies for non-invasive respiratory support. They were given a clinical vignette to put a context to that and sent the survey. They got like 381 responses, 61 physicians, 173 nurses, and 147 respiratory therapists. They represented about 23 tertiary NICUs across Canada. 24 % of the group had zero to five years of experience, 22% had six to 10, and most of the participants, 54%, had more than 10 years of experience.

So, under “perspective on devices and strategy,” let's start with device exposure and usage. So CPAP was the most commonly used form of non-invasive respiratory support reported by 69% of providers, typically usually at pressures up to eight centimeters of water. NIMV was followed at 21% and usually using peak pressures of about 20 centimeters of water with an IQR of about 18 to 24. Now what's interesting is that they asked how people felt about these pressures. Is it high? Do you feel like you guys are using high pressures? There was some interesting interprofessional differences there. Nurses were more likely to feel unsure or feel that CPAP and NIMV level were too high, especially when the pressures were 10 or higher, or that you reached 22 or more on NIMV. For interfaces, the dominant strategy reported by about 70% of groups was to alternate between the nasal mask and the nasal prongs. But notably, 130 respondents actually preferred a different interface that was typically used in their units. I thought that was interesting, that there's not a great level of satisfaction with the strategy of these interfaces. Altogether, when it comes to the perceived difficulty of non-invasive respiratory support, 62 % of physicians, 45 % of nurses, and 45 % of RTs found that the provision of non-respiratory support was challenging.

In terms of “perspectives on monitoring and managing events,” across all provider groups, there were strong agreement that suboptimal positioning, airway obstruction, and poor interface fit contribute to respiratory events. But again, there were some interesting differences that came up. Nurses and RTs were more likely than physicians to view gastroesophageal reflux and anemia as important contributors to the respiratory instability. Documentation was another pain point. Around 20 % felt that the apnea documentation by nurses were not often reliable, with concerns for desaturation and bradycardia entries. However, there was a near-consensus on what constitutes a reportable event. 88 % agreed that events requiring intervention and 69 % agreeing that those lasting more than 20 seconds should be clearly documented.

Under “perspective on failure solutions,” when asked about re-intubation, 57 % perceived that their unit was highly tolerant to respiratory events before acting
this brought me back to a lot of awkward interactions! 39 % noted that tolerance threshold vary widely depending on the physician on call. Again, something we've all encountered. Interestingly, 44 % of physicians believed that reintubation could often or always be prevented. But nurses, 29%, and RTs, 28%, shared that optimism.

As for “preventive strategies,” there was strong support for protocolizing practices related to noninvasive respiratory support, increasing nurses to patient ratio and RT to patient ratios, and establishing clear criteria for reintubation was supported by 88 % of all respondents. Talking about the value of having in-house neonatologists, 33 % of the doctors felt like this was helpful, but 57 % of the nurses thought this was helpful. Clearly some bias happening there from the doctors! On the topic of training, 17 % rated their non-invasive respiratory support education as neutral or insufficient. Hands-on workshops were supported by 84 % of physicians, 76 % of nurses, and 63 % of RTs.

This study just underscores how complex this practice really is, even though we use it fairly commonly, and how collaborative really it needs to be. It's not just about the technology or the interface. It's about how we apply it. It's about who's at the bedside. It's about what training we get and what protocols we have in place. The authors really emphasize that we need better scientific evidence on the application, maintenance, and monitoring. We need interprofessional collaboration to understand behavioral and organizational influences and standardized guideline and harmonized training to reduce variability. So it's a team-based approach, and this was definitely a good read.


Daphna Yasova Barbeau, MD

It's nice to see that, especially when lots of our different staff types and members are included in the evaluation.

I had a really quick one, which will hopefully allow you to do your last one. This is in JAMA Pediatrics called “Prenatal Cannabis Use and Neonatal Outcomes: A Systematic Review and Meta-Analysis.” Lead author is Jamie Lo, senior author is Devan Kansagara. They wanted to see if cannabis use in pregnancy was associated with neonatal outcomes. This was a meta-analysis, and we don't like to do a lot of review papers like this, but even in their background, they said it's not ethical to conduct randomized clinical trials of prenatal cannabis use, given our concerns about its safety. So observational studies are the best thing that we have.

They used 51 studies in 69 publications. They included 21,146,938 infants! They had done this meta-analysis once, and now they're doing it a second time using eight more studies since the original publication date, adding a million new participants. The other thing that was different about this study was they wanted to also look at, was there a dose response? Some of the studies provided information about dose response, and they extracted the heaviest users from those studies. For instance, they say if a study reported outcomes for cannabis monthly or less, weekly, or daily, then that daily use group was included in the heavy use group.

They found 24 studies analyzing this association between prenatal cannabis use and low birth weight infants in their meta-analysis. They found an increase odds of low birth weight infants in those with prenatal cannabis use and odds ratio of 1.75. In their dose response, they found that in those with heavy cannabis use, the odds of low birth weight increased to an odds ratio of 2.36; certainty of evidence is moderate.

Then they looked at prenatal cannabis use and preterm birth. Odds of preterm birth were increased for those using cannabis in pregnancy and odds ratio of 1.52; for those with heavy use, a combined odds ratio of 1.95. Certainty of evidence is moderate.

Then they wanted to look at perinatal mortality. They found an increased odds of perinatal mortality with cannabis use and odds ratio of 1.29. There were three studies that had this specification about heavy use, but they actually found that the ads were no longer statistically significant in this three studies that included heavy use.

They found 21 studies that looked at cannabis use and small for gestational age. The odds of SGA increased with prenatal cannabis exposure was an odds ratio of 1.57; for those with heavy use, an odds ratio of 1.63. Certainty of evidence moderate.

I was hoping there would actually be more perinatal outcomes, but these are the biggies. I think that we are seeing more marijuana use. I always like to take a look at the marijuana papers when they come in.


Ben Courchia, MD

For sure. I haven't gotten off my initial of stance on this. Legalization doesn't mean beneficial. Just because you have more access to something, it doesn't mean that it's good for you. We know the effects of marijuana, unfortunately, and the legalization doesn't change that. Just have to be careful.


Daphna Yasova Barbeau, MD

It's like so many other things. Alcohol, cigarettes, etc.


Ben Courchia, MD

Yeah. Alcohol, cigarettes. Absolutely. Even illegal drugs. The political context doesn't change the substance itself. The substance is what it is.

We're close to the end. I'm not going to extend this journal club too much, but there was an article that came out in the Archives of Diseases in Childhood that caught my eye and it was less impactful than I thought. It's “Near-infrared spectroscopy during respiratory support at birth: a systematic review.” First author is Vix Monnelly and it's coming from the ILCOR (International Liaison Committee on Resuscitation Neonatal Life Support Task Force). This is a big group publishing on a potentially an innovative approach to management in the delivery room. They're talking about perinatal transition and compromised cerebral oxygenation that may lead to morbidity and mortality.

The question is, should we start using NIRS in the delivery room to guide respiratory management? There are some issues, obviously, with the potential randomization. This is just a systematic review and meta-analysis based on Ovid Medline, Embase, and Cochrane Central.

They looked at newborn infants of all gestation. They were looking at studies that looked at any newborn receiving CPAP or NIMV during resuscitation. They looked at the intervention from studies that would look at CRS02 (cerebral oxygen saturation) with a dedicated treatment guideline they were looking at outcomes that involved survival without neurodevelopmental impairment, survival alone, severe IVH, PVL, etc. They reviewed close to 600 articles, and they only found three articles reporting the outcomes from two randomized trials. What they're saying is that no data was found for survival without neurodevelopmental impairment, so they could not exclude benefit or harm from delivery room monitoring of cerebral regional saturation for survival, severe IVH, PVL, or cerebral regional saturation below the 10th percentile.

This paper just highlights that the limited evidence can't really exclude either benefit or harm. If you want to read the paper, you're more than welcome to. I thought, is this the paper that's going to lead to a revision of our approach to resuscitation? But the data is just not there yet, so not much to add.


Daphna Yasova Barbeau, MD

Sorry, I was muted. We’ve longest day we've had in a very long time. But should I do one more? And then we can be done. I'm going to try to review this quickly. You know, I'm not very good about that, but this is in Journal of Perinatology. This is coming to us from our colleagues in Japan and it's entitled “Neurodevelopmental Outcomes at Age Three Years of Preterm Infants Born at 22 to 31 Weeks Gestation.” I'm not taking the smallest of papers, but I didn't want it to linger. I think people are always wondering and the paper came out and I just wanted to make sure we do it.

They were looking at infants born before 32 weeks and they were actually looking at this group from 2012 to 2014 (almost a decade ago) from their improvement of NICU practices and team approach cluster trial. This was the INTACT trial (Improvement of NICU practices and Team Approach Cluster randomized controlled Trial). They were looking for survival without moderate or severe neurodevelopmental impairment. The INTACT trial was this quality improvement program in Japan that didn't have the effect that they had hoped it would in changing outcomes for babies who were enrolled in this package of interventions, but they are just now reporting on the neurodevelopmental follow up.

They had 40 NICUs that were allocated to the intervention. Initially were 19 NICUs and 21 control NICUs. I'm actually going to skip telling you any more about the original INTACT trial. I'm going to just tell you about the current trial [that was a secondary analysis]. They combined both the intervention and control groups to get this cohort. They wanted to look at infant survival at age three. They defined moderate neurodevelopmental impairment as neurologic disability with one or more of following conditions: cerebral palsy with a GMFCS level of two, Kyoto Scale of psychological development less than 70 (developmental quotient less than 70 is equivalent to a Bayley three score less than 85); visual impairment requiring correction, but not been fully corrected; and some other things about vision.

Ben Courchia, MD

Yeah, it feels more like a Bayley 2 almost where they get this one single score. On the Bayley 3, you would have subcategories, you could break that down. It feels almost like this MDI that you used to have on the Bayley 2.


Daphna Yasova Barbeau, MD

Exactly. Ok then severe NDI included: cerebral palsy with a GMFCS level 3-4; Kyoto Score less than a 55; blindness or ability to perceive light only; impaired hearing that's not correctable; and any individual residing in a facility for children with severe physical disabilities was considered to have severe neurodevelopmental impairment.

They have these beautiful tables by week of gestational age, which we will get to in just a second. They had 2,722 infants evaluated at NICU discharge. They had a total of ~2,360 infants which were 90 % of the survivors that were able to attend the follow up visit and were included in the analysis of the outcomes. For the group, I wanted to tell you there are 43 22-weekers, there are 152 23-weekers, there are 197 24-weekers, 220 25-weekers, 280 26-weekers, 350 27-weekers, 370 28-weekers, 400 29-weekers, almost 400 30-weekers and 284 31-weekers.

We can talk a little bit about the clinical characteristics and morbidities. A higher percentage of SGA infants, so a birth weight below the 10th percentile was observed with increasing gestational age. Antenatal steroids were observed for 44 % at 22 weeks and greater than 60 % at the other weeks. So most of their 22-weekers still didn't get antenatal corticosteroids. A higher incidence of chronic lung disease requiring continued oxygen use and respiratory support at 36 weeks. IVH, sepsis, and NEC was noted with decreasing gestational age. The incidence of PVL was the highest at 24 weeks gestation, only 5%, followed by 4% at 23 weeks and 3.5% at 28 weeks. The overall NICU mortality rate was 3.5%, 5 % for all infants. The mortality rate decreased with increasing gestational age, ranging from 28 % at 22 weeks to 0.4 % for those at the 31 week gestation mark.

They did neurodevelopmental outcomes at age three and the composite outcome of survival without moderate or severe neurodevelopmental impairment. There's a lot of information here. But a higher percentage of SGA children was observed with increasing gestational age, which was interesting, 0 % in the 22-weekers and 56 % at 31 weeks gestation. Infants with cerebral palsy of GMFCS level 2 to 4 was 4.7 % and was highest in the infants born at 23 weeks gestation, 10 % of infants in the 23 weeks. All infants born at 22 weeks gestation with cerebral palsy (10 % of 22 weekers) had a higher level of cerebral palsy, GMFCS greater than or equal to level three. The percentage of children with cognitive delay decreased over gestational age. 54.8 % for those born at 22 weeks to 10.8 for those born at 30 weeks. Additionally, the proportion of children with severe cognitive delay correlated again with gestational age. But the proportion of children with this borderline cognitive level was similar in children born at 22 to 27 weeks gestation. 12.9 % of children born at 22 weeks and 24 % of those at 23 weeks had what was considered “normal” cognitive development. Two infants born at 25 and 29 weeks had blindness or light perception only. The percentage of visual impairment requiring slight correction but not fully corrected was 2%. Two infants born at 26 and 27 weeks had severe hearing impairment that was not correctable.

Among the infants with the full evaluation, 13 % had moderate neurodevelopmental impairment and 7.8 % had severe neurodevelopmental impairment. The survival without moderate or severe neurodevelopmental impairment for the whole group was 79%. But this was higher, obviously, in the higher gestational ages. Survival without moderate to severe neurodevelopmental impairment was 45 % at 22 weeks, 57 % at 23 weeks, and 88 % at 31 weeks.


Ben Courchia, MD

Yeah, but those numbers at 22 weeks were quite low in terms of the census that they had.


Daphna Yasova Barbeau, MD

They had 31 surviving infants at 22 weeks. The babies with no cerebral palsy was 87%. But all the babies who had cerebral palsy, this 10%, almost all of them had severe cerebral palsy. 3% had mild, 10% had severe, and 87% had no cerebral palsy.

Of the 22-weekers, 12.9 % had no cognitive delay and 32 % had borderline cognitive delay. 32.3% were delayed, and 22.6% were severely delayed.

I thought it was interesting. They have these nice graphs that look at by gestational and week of life, looking at survival, moderate and severe neurodevelopmental impairment. Again this was from a cohort almost a decade ago. They've been doing the 22-week thing longer than most centers in the States. I'd love to see an updated cohort. But I these outcomes are consistent with our United States cohort, potentially better than our United States cohort.

Ben Courchia, MD

It's an interesting story, because we've always been very interested by what the Japanese have been doing because we know they have good outcomes, so we've been following that closely. For the first edition of Delphi, we wanted to take that opportunity to bring someone from Japan to talk to us. We were able to. Dr. Namba from Japan came and gave a presentation on what is the approach to babies on the edges of viability. It cost us a fortune, as you can imagine. We were a small conference! The video is now on YouTube. It's important to share what's so great about what they do. It’s great we've been able to engage with this community and talk to them about their protocols and their guidelines on.


Daphna Yasova Barbeau, MD

He does provide from their institution their outcome data during that talk. Definitely for people to take a look.


Ben Courchia, MD

One of the things that he's mentioned is that he has said that their outcomes are phenomenal for survival, right? They have a very interesting approach. He mentioned how their survival is phenomenal, but in his presentation, he said the next frontier is survival that is free of neurodevelopmental impairment. It's not surprising to see those numbers and it's important for people to understand that they are a little bit ahead of us, in the sense that they're doing very, very well when it comes to survival, and now they're going to start modifying their protocols so that they can optimize for neurodevelopmental outcome. But we do have to take everything with a grain of salt, because I don't know the Kyoto developmental scale. I don't know if that translates one-to-one with a Bayley, but some other things are a bit more objective, like the CP diagnosis. That is fairly similar in terms of how we diagnose that. There's still some differences. But think it's nice when you're able to blaze your own trail. We have the Scandinavian approach, we have the Japanese approach, so I'm always interested in seeing what the tenets of these different methods are.


Daphna Yasova Barbeau, MD

All right, buddy. I'm done.


Ben Courchia, MD

Well, thank you very much. This was a great journal club, and we will see you all next time. Bye.




 
 
 

ComentĂĄrios

bottom of page