#458 - [Journal Club] - 📌 Does holding during cooling change NICU outcomes in HIE?

Parents of babies undergoing therapeutic hypothermia often wait days before they can hold their child, and many never get the chance during cooling at all. This week on Journal Club, Daphna reviews a new Journal of Perinatology retrospective cohort from two level IV NICUs, 379 infants with HIE, where 28% were held during cooling. Held infants reached full oral feeds sooner, were more likely to go home on breast milk, and had shorter stays, but they were also less sick to begin with. Ben pushes on confounding, on the exclusion of infants who died, and on what equipment would make holding routine.

----

Association of parental holding during therapeutic hypothermia and NICU outcomes for infants with hypoxic-ischemic encephalopathy. Nguyen TT, Glass HC, Chan N, Taketa E, Pineda R, Cornet MC, Miller MJ.J Perinatol. 2026 Jun 22. doi: 10.1038/s41372-026-02753-3. Online ahead of print.PMID: 42332041

Transcript

Ben Courchia, MD (00:00): Hello everybody, welcome back to The Incubator Podcast Journal Club. It is Tuesday and we are back today looking at some more papers. Daphna, I think it's your turn today to talk about the research you've identified.

Daphna Yasova Barbeau, MD (00:15): Yeah, it's my turn. This is an article out of the Journal of Perinatology, lead author Tiana Wynn and senior author Matthew Miller. I know families who have undergone therapeutic hypothermia have been awaiting this paper. It's entitled "Association of Parental Holding During Therapeutic Hypothermia and NICU Outcomes for Infants with HIE (Hypoxic Ischemic Encephalopathy)." It was a retrospective cohort study of infants diagnosed with HIE and admitted to UCSF Benioff Children's Hospital for therapeutic hypothermia between March 2017 and February 2024. They looked at whether the infant was held during the course of therapeutic hypothermia. And why is this controversial? Well, babies have a lot of lines on them, central lines.

Ben Courchia, MD (01:21): Are you seriously asking that? I mean, there are a lot of nurses who say—well, I'll tell you why it's controversial.

Daphna Yasova Barbeau, MD (01:27): That's true for a lot of our babies in their first few days of life in the NICU. But I think one of the things people have worried about in the past is whether holding will change the outcome of the cooling itself—will we be able to maintain the temperature? There have been previous studies in just the last handful of years showing that the blankets we have today are able to manage temperature fluctuations during holding. So that's exciting.

To give you more detail on the primary variable of interest, they also looked at when the first hold occurred, what day of life, and the frequency of holding. They looked at a number of infant characteristics and clinical factors, including severity of encephalopathy using the Sarnat exam, some lab values, MRI results, history of seizure, respiratory support, medication need, and total number of intubation days. Did the babies need a gastrostomy (G) tube? They were really looking at severity of illness. The NICU outcomes included time to full oral feeding, day of life at full oral feeding, receipt of breast milk at discharge, and length of stay—all things we're looking at all the time when we're thinking about new interventions in the NICU.

So what did they find? They had a total of 440 infants with HIE who received therapeutic hypothermia at the study sites during the study period—two level 4 NICUs. They excluded infants who died during the birth hospitalization and infants transferred to another NICU—61 infants excluded—resulting in a final sample of 379 infants, mean gestational age 39.4 weeks, mean birth weight 3.3 kilos. Of the 379 infants, 28% were held during therapeutic hypothermia. The proportion of infants held during cooling increased over time across the years of the study, particularly in the more recent years, with some variation observed between the two sites.

Among infants held during therapeutic hypothermia, the time to first hold was 1.4 days, compared to about four days for infants who were not held—which I think in and of itself is a success. A lot of times these babies get rewarmed, and we still miss the opportunity to have their parents hold them right after rewarming. In addition, infants held during therapeutic hypothermia had a higher daily frequency of holding during their hospitalization compared to those not held—three times a day versus 2.5—which was statistically significant.

Now, which babies were being held? What characteristics were associated with infant holding? I don't find them particularly surprising. Infant holding during therapeutic hypothermia was more common among infants with higher five-minute Apgar scores—an average of five versus four—and more common in babies who had private insurance, 68% versus 32%. There was also a difference by site. Race and ethnicity were significantly associated with holding as well, with higher rates among infants identified as white or multiracial, and lower rates among those identified as Hispanic, Asian, or Black. There were no associations between holding during therapeutic hypothermia and gestational age, birth weight, sex, pH, base excess, presence of brain injury on MRI, or history of seizures. So it seemed like babies who were less sick were more likely to be held, along with other factors related to family comfort and advocacy in a medical setting.

What were the NICU outcomes? Infants held during therapeutic hypothermia had more favorable NICU outcomes compared to those who were not held. Again, I'll underscore that in the initial analyses this appeared to be a group of babies who were less likely to be intubated and had higher Apgar scores. But infants who were held achieved full oral feeding earlier—a median of 2.6 days versus 3.6 days. They were more likely to receive breast milk at discharge—92% versus 78%—and they had a shorter length of stay, a median of eight days versus 11 days.

In the unadjusted analysis, infants who were held achieved full oral feeding earlier than those who were not held, whether measured from birth or from initiation of oral feeding. So regardless of how it was measured, the babies who were held got to full feeds sooner. Measured from initiation of oral feeding, the unadjusted median time was 2.6 days for held infants compared to 3.8 days among those not held. [Note: see flag below regarding a discrepancy with the earlier-stated 3.6-day figure.] When measured from birth, the median day of life at full oral feeding was 5.9 days among held infants and 8.4 days among those not held.

In the adjusted analyses—which controlled for severity of encephalopathy, number of intubation days, insurance, race, ethnicity, admission year, and site—holding during therapeutic hypothermia was still associated with a faster rate of achieving full oral feeding from initiation of oral feeding, with a hazard ratio of 1.4, which was statistically significant. It was also associated with a faster rate measured from birth, with a hazard ratio of 1.6, also statistically significant, compared to infants who were not held. Infants held during therapeutic hypothermia were also two and a half times more likely to receive breast milk at discharge, and they had an average of 4.9 fewer days of hospital length of stay than infants who were not held.

They also ran sensitivity analyses including or excluding race, ethnicity, and insurance from the adjusted models, and the associations remained statistically significant. So I think this is another piece of information underscoring that holding isn't just a nice thing to do—it's not just nice for the parents or the babies, it may actually improve NICU outcomes, like getting babies home sooner. Thoughts?

Ben Courchia, MD (09:35): Yeah, I have some thoughts and some issues. I don't know what to make of the fact that—as you mentioned—these are babies who are more likely to be held because they're likely less sick. So you wonder, are they doing better because they're just less sick at baseline, or because they got held?

Daphna Yasova Barbeau, MD (09:57): They did run the adjusted analyses to account for that.

Ben Courchia, MD (10:03): Sure. And the point is, I almost don't care about that, because I don't need these outcomes to tell me it's good for parents to hold their babies during therapeutic hypothermia. To me, it's one of these interventions where I don't need the outcomes—we know from being physicians that this is good for babies and parents. Now, the real question I have for you—full disclosure, we have no vested interest in anyone providing technological solutions for HIE—do you know of any new tools today that make holding babies during cooling more achievable?

Daphna Yasova Barbeau, MD (10:53): Interesting. Well, first, let me respond to your first comment. We say babies should be held—absolutely. But I do agree that there are nurses at the bedside who say there are risks to holding babies: line malposition, for one. Fine. So I think having this data is important, so we can say yes, there are risks, but there are also concrete benefits we need to discuss.

Ben Courchia, MD (11:10): Agreed. I'll come back to that—that's my final question for you.

Daphna Yasova Barbeau, MD (11:20): To answer your question, there are definitely cooling interfaces made specifically for babies—that's a start. There's the infant blanket, which we know impacts temperature regulation and control. There are also different types of wraps, and that may come down to hospital preference, but I think that impacts whether nurses feel comfortable doing the holding. That may just mean they need additional education on how to do it safely.

Ben Courchia, MD (12:06): So you mean there are blankets that wrap around the baby? Because right now the one I know of is more like a mat—it has water running through it, you put it on the bed, you put the baby on it, and that's kind of it.

Daphna Yasova Barbeau, MD (12:14): Right, that's a mat. But there are some that are wraps—you wrap the arms and legs, some are adhesive. They haven't had a head-to-head comparison for holding, but potentially those are opportunities.

Ben Courchia, MD (12:33): Do you think, in your experience, wrapping the arms and legs makes it easier?

Daphna Yasova Barbeau, MD (12:42): Yeah, I think so, for sure. That's how nurses are used to handing babies to parents—kind of swaddled. We know cooling itself causes some irritability, and swaddling potentially makes that more comfortable for everybody. To my knowledge, there's no study specifically on swaddling babies who are being cooled, so that's potentially something that could be looked at. Some other technological advances relate to how we secure our lines. If lines are the reason we're not holding babies, maybe we need to look at why they're not secure enough, and whether there are interventions to secure them so that's no longer an exclusion criterion—both for this group of babies and for everyone trying to get ELBW (Extremely Low Birth Weight) infants held sooner. Are there ways to make endotracheal tubes and lines more stable so we can meet these goals? I think that's a major challenge. Parent education and comfort matter too.

Ben Courchia, MD (13:57): My other question was about the exclusion criteria—specifically excluding babies who passed away. What if the babies who were held had a higher mortality rate? That would be a significant finding. Not that it happened—I don't know the data, and I'm not familiar with the authors or their work—but do you think that's something the study could have benefited from including?

Daphna Yasova Barbeau, MD (14:31): Yeah, absolutely. That's an ongoing discussion about excluding babies who die. This was a retrospective study, so obtaining consent from bereaved parents is complicated, but not impossible. In our discussions with bereaved families, many of them are looking for ways to honor their babies or to provide information that could help babies who will have experiences like theirs. I actually think we do bereaved families a disservice by not approaching them for research, because I think many of them would say yes. Especially here, where we know there are benefits to holding that we can't always quantify—what if we could? Looking at those babies may have made this an even more drastic finding. And if babies are going to die, I wonder about the value of providing them this intervention, or letting families have it.

Ben Courchia, MD (15:52): Love that point—very good point. All right, those are all my questions. Thank you for running through it. Great paper. We're going to wrap up for now, and we'll see you all tomorrow.