#453 - [Journal Club] - 📌 Should We Treat the PDA Based on Size Alone? (SMART PDA Trial)

In this Journal Club, Ben and Daphna dig into two new papers on PDA management in our smallest patients. First, the SMART-PDA pilot RCT from Souvik Mitra and colleagues, which uses comprehensive hemodynamic screening to selectively treat high-volume shunts in infants born before 26 weeks, and whose striking Bayesian signal for reduced pulmonary hemorrhage and NEC stopped the trial early. Then a companion JAMA Network Open comparative effectiveness study across four pharmacotherapy regimens. Along the way, Ben shares hemodynamics pearls from his Montreal training: why left ventricular output, LA:Ao ratio, and transductal velocity matter more than PDA diameter alone.

----

Selective early medical treatment of the patent ductus arteriosus in extremely low gestational age infants: a pilot randomised controlled trial (SMART-PDA). Mitra S, Hebert A, Castaldo MP, Disher T, El-Naggar W, Dhillon S, Alhassen Z, Koo J, Katheria AC, Hyderi A, Kumaran K, Ting J, Surak A, Larocque J, Pepper D, Hornberger L, Makoni M, Weisz DE, Jain A, Bacchini F, Cameron-Nola AJJ, Hatfield T, Dorling J, McNamara PJ, Thabane L.Arch Dis Child Fetal Neonatal Ed. 2026 May 18:fetalneonatal-2026-330462. doi: 10.1136/archdischild-2026-330462. Online ahead of print.PMID: 42150872

Pharmacologic Therapies for Patent Ductus Arteriosus in Extremely Preterm Infants. Mitra S, Jain A, Ting JY, Ben Fadel N, Drolet C, Abou Mehrem A, Soraisham AS, Jasani B, Louis D, Lapointe A, Dorling J, Khurshid F, Hyderi A, Kumaran K, Toye J, Harabor A, Weisz DE, Stavel M, Morin A, Bhattacharya S, Lalitha R, Afifi J, Augustine S, Castaldo MP, Hatfield T, Su YC, Shah PS; Canadian Neonatal Network Investigators.JAMA Netw Open. 2026 Jun 1;9(6):e2617477. doi: 10.1001/jamanetworkopen.2026.17477.PMID: 42262753 Free PMC article.

Transcript

Ben Courchia, MD (00:00.737)
Hello, everybody. Welcome back to the Incubator Podcast. We're back today for an episode of Journal Club. Daphna, good morning. How are you?

Daphna Yasova Barbeau (00:08.75)
Good morning, good morning. You have been jet setting, huh? It's nice to have you back in town.

Ben Courchia, MD (00:12.619)
Yeah, I did. We went on vacation, the one week of vacation that we take, and went to France. It was a lot of fun. And from there I went straight to Montreal for my hemodynamics training, which was something new.

Daphna Yasova Barbeau (00:29.388)
Yeah. What did you learn? What was new about it? You've been doing this for a year now.

Ben Courchia, MD (00:37.045)
A year now. Well, we're going to talk about it today, actually, because I think that with these two papers that I'm going to review today, I'll be able to shed some light on some of the hemodynamics things I've learned. I wanted to maybe just do a few announcements. I believe that when this episode is going to air,

Daphna Yasova Barbeau (00:39.586)
Okay.

Ben Courchia, MD (01:04.833)
Well, let me just check when this episode is going to air.

Daphna Yasova Barbeau (01:06.958)
Let's walk it back a second.

Ben Courchia, MD (01:09.717)
Let's walk it back a second. I just want to make sure that I'm not giving people the wrong information. So when this episode is going to air, there will be three days left to submit your feedback on our website for the annual anniversary of the podcast. And then you can be entered to win a whole set of prizes and so on. So make sure you do that. I think that this is quite nice.

Daphna Yasova Barbeau (01:17.322)
No.

Ben Courchia, MD (01:39.213)
In terms of the other thing that I wanted to mention, I've seen a lot of doctors starting to use a lot more Substack, which is a platform I don't fully understand. I understand it's kind of like—okay, I'm going to upset so many people—but it's kind of like a blog, and then you can subscribe to people's Substack. I've seen it used a lot with journalists or people who are in the news. But one of our

Daphna Yasova Barbeau (01:50.038)
Yeah.

Ben Courchia, MD (02:08.343)
Dear friends, Nick Embleton from the UK has started his own Substack. He has one on NICU nutrition and one on baby loss. So check him out. I think he's always been a strong voice for our community. He's a very bright mind, and he has a lot of insight when it comes to both nutrition and infant mortality. So you can go on Substack and look for him.

And subscribe. That's really it. So these were my two announcements for today. Anything else that I forgot? Otherwise, we'll deal with that tomorrow. Okay.

Daphna Yasova Barbeau (02:49.452)
No, I think that's it. I'm excited to get this hemodynamics primer today.

Ben Courchia, MD (02:56.517)
Yeah. So today I'm going to get to a paper that I've been meaning to review. It's published in the Archives of Disease in Childhood, Fetal and Neonatal Edition. It's authored by Souvik Mitra, who I guess is taking up a lot of space on our podcast these days, but that's not because of a lack of merit. He's very prolific. He's doing phenomenal work.

Daphna Yasova Barbeau (03:12.499)
He's quite busy, this guy. Very prolific. I don't know how he gets it all done.

Ben Courchia, MD (03:22.593)
But this is the selective early medical treatment of the patent ductus arteriosus in extremely low gestational age infants, a pilot randomized controlled trial. These are the results of the SMART-PDA trial. This is a very important paper, I believe, for the discussion on the PDA. And, to be honest with you, it's not an easy paper to understand. It's quite a complex design. And I think that's the—

My preface is basically based on the fact that Souvik and his team have tried to address some of the gaps in what we know and how we study the PDA. So once you understand it, it makes sense and it's easy, but I'm hoping that I can break it down for the audience. So I'm not going to give you too much background on the PDA. Most common cardiovascular condition affecting our preemies, pretty bad downstream effects of the PDA—NEC,

Bronchopulmonary dysplasia, et cetera, et cetera. Now, there are large randomized pragmatic trials that have not really demonstrated the benefit of treating the PDA. And what we use to treat the PDA are the cyclooxygenase inhibitor drugs: indomethacin, ibuprofen, acetaminophen. And for babies that are born extremely prematurely, the micropreemies, which is what we're talking about in this paper, the babies less than 26

weeks, there's still a lot of desire from a lot of institutions to treat these infants. And I think this is where the contradiction lies, right? We keep showing data that we're moving away from treatment, but for this population, the question is, well, maybe this population benefits from it. Interestingly enough, the paper mentions a survey of the Canadian Neonatal Network that shows that about two-thirds of this population

in Canada receives pharmacologic treatment for their PDA. Now, given the potential adverse effects of the medications that we have available and their waning effectiveness as we go up in postnatal age, there is a controversy as to whether doing a selective early treatment of a high-volume PDA shunt can improve clinical outcomes in this smallest infant group at the highest risk of

Ben Courchia, MD (05:47.374)
PDA-attributable morbidity. And to their knowledge, they said there's no RCT that's been conducted exclusively on this micropreemie population of less than 26 weeks that employs a comprehensive hemodynamic screening strategy to specifically select and treat high-volume PDA shunts. And I think that's very important. We have a lot of data, most notably, I would say, from the University of Iowa,

where they have a hemodynamic program that helps them address the PDAs of their patients. And it's a report of their practice, but these are not randomized controlled trials. They've adopted this practice. So this is where the distinction comes in. And for people who must know, Dr. Patrick McNamara, who leads the team at the University of Iowa, is one of the senior authors of this paper. So

We'll talk a little bit about the trial. So the selective early medical treatment, abbreviated the SMART-PDA trial, is a multicenter, open-label, parallel-design pilot RCT conducted in seven Level III and Level IV centers that all have an established neonatal hemodynamics program that offers targeted neonatal echocardiography.

Four were in Canada, three were in the US. One point about the trial and how it was ultimately analyzed: what they did is basically conduct a Bayesian analysis throughout the study to inform whether the SMART-PDA approach was the right one.

And the Bayesian analysis was so striking that they basically had to stop the trial and just publish these results as is, as a standalone trial, because of the unwillingness of multiple participating centers to randomize their micropreemies to the control group, meaning the

Ben Courchia, MD (08:11.028)
non-treatment group. So I think this is important. And I'm going to tease this a little bit, because I'm not telling you what this Bayesian analysis has shown. I think that when I tell you, you will understand why some people said, no, I'm not doing that. So in terms of the participants, we're talking about infants born below 26 weeks of gestation who were enrolled, obviously with consent from the parents.

Daphna Yasova Barbeau (08:12.846)
Mm-hmm.

First.

Ben Courchia, MD (08:37.408)
Infants with antenatal or postnatal major congenital anomalies, congenital heart disease, or babies that were on the palliative care track were obviously excluded from enrollment. All babies who were enrolled got an echo done within the first 72 hours of life. And infants with a PDA, any shunt severity documented on the initial screening, regardless of prior exposure, were eligible for randomization. Infants in the SMART-PDA

protocol got an echo every 72 hours for up to seven days of age to categorize the PDA shunt severity. I'm going to describe to you what the SMART-PDA algorithm is, because I think this is where this trial really adds a lot to the conversation. We've been discussing how we can categorize what's known as a hemodynamically significant PDA.

And I think they're doing a very good job at describing that. The way they're doing this is that they have three categories, putting the PDA into buckets. You can have mild, moderate, severe—and we have mild, moderate, severe categories for both clinical signs of the PDA and echocardiographic markers. So the clinical signs of a mild PDA are that you require less than 30% oxygen and that your mean airway pressure is less than eight centimeters of water.

I'm going to keep going through the clinical criteria and then I'll tell you about the echo criteria. The moderate PDA is needing 30 to 50% oxygen and requiring a mean airway pressure of 8 to 12. There is reduced urine output and systemic hypotension, which is defined as a blood pressure less than the gestational age in weeks at birth, which we know has pitfalls, and that's by convention. And then we have the severe

clinical PDA, which is requiring more than 50% oxygen, more than 12 centimeters of water of mean airway pressure, profound or recurrent pulmonary hemorrhage, acute renal failure, and hemodynamic instability to the point that you're requiring more than one cardiotropic agent. So, so far, so good, Daphna. This is pretty straightforward. It makes sense.

Ben Courchia, MD (10:57.384)
Then let's talk about echo findings. I feel like before I did hemodynamics, I did not really understand the true assessment of a PDA. And so I think it's always good to buff up on that particular aspect. So the point here is that the size of a PDA does not necessarily mean that the PDA is hemodynamically significant. So a mild PDA has a size of less than 1.5 millimeters.

A moderate PDA is 1.5 to 2.5 millimeters, and a PDA above 2.5 is considered severe. But that's not sufficient. It is also about what else that PDA tells us. So for a mild PDA, we're looking at a left atrium-to-aortic root ratio of less than 1.5 and a left ventricular output of less than 200. So what could be confusing is, why are we looking at the left atrium?

Why are we looking at the left ventricular output as a way of assessing the PDA? So this is my pearl from hemodynamics that you can take home. The point here is that the output of the left ventricle is the direct product of pulmonary flow. Right? So when you talk about Qp:Qs, the output of your left ventricle is

the direct consequence of Qp. And when your PDA shunts from left to right, you are donating a lot of your systemic output to the pulmonary flow, to the pulmonary vasculature, which means that all this flow is eventually coming back to your left heart. And that's what's causing your left atrium to be flooded because of that excess pulmonary flow.

which leads to your left atrium enlarging, which is why your LA:Ao ratio increases. So that's why an LA:Ao ratio of 1.5 or more is abnormal, because it signals that your left atrium is enlarged from all this returning blood volume. And if your left ventricular output is high, you might think that's good—my left ventricle is working well. But basically, what that means

Ben Courchia, MD (13:18.22)
is that your left ventricular output is high because it's dealing with a lot more blood flow than it needs to. So when you look at this, a left ventricular output of less than 200 is good. So that's a mild effect. A left ventricular output of 200 to 400 mL per kilo per minute, that's moderate hemodynamic significance of the PDA. And a severe would be a left ventricular output of more than 400. Does that make sense?

Daphna Yasova Barbeau (13:45.539)
Yeah, I think that's really helpful, right? So backflow into the left atrium, more flow now moving into the left ventricle. Yeah.

Ben Courchia, MD (13:54.016)
Exactly. The other thing that they looked at was transductal peak velocity, so the velocity across the duct. And obviously your severity went up based on an increasing level of transductal peak systolic velocity—meaning, sorry, decreasing peak systolic velocity. Because what you have to think about is that if the velocity across the PDA is high,

The way Gabriel explains it is that he mentions a hose where you hold your thumb to the end of the hose. If it's just squirting very fast, that means there's restriction of the PDA. But if the PDA is wide open, the flow is not going to be going super fast. It's more going to be like flow going through an unobstructed outlet. And so that's—

Daphna Yasova Barbeau (14:44.163)
Yeah, and I think it's hard sometimes for people to understand. The flow volume is increased, but the flow velocity or the rate of flow is decreased. Yeah. But the problem is the flow volume.

Ben Courchia, MD (14:56.193)
Exactly.

Ben Courchia, MD (15:02.529)
Exactly. You got it. And then the last piece of the assessment is diastolic flow in the descending aorta.

So obviously in diastole, if you have a significant shunt, you will get steal, which means that the flow in the descending aorta during diastole will be reversed or absent.

Daphna Yasova Barbeau (15:26.563)
Sure, that makes sense. Reversal on the aorta, bad.

Ben Courchia, MD (15:31.352)
Yeah, exactly. Okay, so let's go back to the procedure. So pharmacotherapy, when it was indicated, was provided in the form of either ibuprofen, 10 mg per kilo once, followed by 5 mg per kilo times two every 24 hours. The administration route was decided by the team. And if there was a contraindication, then they could use acetaminophen given as 15 mg per kilo per dose q6 for three to seven days. Again, IV or PO. Infants randomized to the control arm—

So the control arm here did not undergo any further echocardiography or PDA treatment, regardless of their clinical signs in the first seven days, after which the decision on PDA management was made at the discretion of the treating physician. So basically, the control arm is no treatment for the first seven days. And I'm going to put this in context, by the way, because as you remember, the recent recommendations from the American Academy of Pediatrics have been

that we don't want to treat the PDA in the first two weeks. So we're doing something that doesn't follow that recommendation specifically, but again, it is guided by echo and it's in a population that we don't have a lot of data on. So what is the management plan? The plan is that you will treat babies if they have moderate signs of a PDA on clinical and echo findings, or if they have severe signs of a PDA on echo and anything

on the clinical. So if you have severe echo findings, you treat no matter what the clinical says. If you have moderate echo findings, you treat only if the clinical aspect is moderate as well. Make sense?

Daphna Yasova Barbeau (17:07.001)
Got it.

Ben Courchia, MD (17:07.831)
All right. In terms of outcome measures, the outcome measure was the proportion of eligible infants enrolled, right? Which is very bizarre when I read it, because it's not what we're used to. We used to say mortality, some form of clinical outcome. But here the idea was, can we actually do a trial like this? And it goes back to a lot of the issues that people have had running PDA trials, which is that you put people in groups and they cross back and forth, no matter what the investigators want. You have kids

who are in a group where it says, hey, you're not going to treat, and then they get treated. And then you're like, well, what do I do with that? And then you try to maybe use intention to treat or something like that, but it pollutes your data quite substantially. And so here the idea was, could we potentially do that? Is that going to work? Is this model going to work? Obviously, they did have secondary clinical outcomes that included mortality, procedural PDA closure, PDA pharmacotherapy, open-label rescue pharmacotherapy, BPD,

postnatal corticosteroids, pulmonary hemorrhage, duration of invasive mechanical ventilation, IVH, PVL, and so on and so forth, ROP, NEC, et cetera. They also pre-specified safety parameters that I will let you review. I feel like I'm extending myself too much, but I want to get into the statistics. This is one of these papers where you must understand the stats. And so what they did is that

they used this Bayesian framework and not the usual p-value approach. And that was deliberate, because it's a pilot trial like this one, and they have about 100 babies. So it was not really powered for statistical significance. So the Bayesian analysis answers the question we actually cared about, which is: given the data, what is the probability that the SMART strategy—SMART being the approach of the paper—is better? And so

instead of looking at whether it's significant or not for every outcome, they treated anything above 80% as a meaningful signal. What they then did is this thing called the win ratio, which is how they built their headline composite outcome. And a traditional composite outcome treats every component as equal. So death is the same as, say, ROP. But in the win ratio, you rank the outcomes by importance, which is also why it's interesting, because you had

Ben Courchia, MD (19:32.281)
in this paper one of the authors, Fabiana Bacchini, who we've interviewed on the podcast, who represents parents. And I think this is key in this paper: what is the outcome that matters to parents, and how does that influence how we treat? And so in this case, they had these ranked using this win ratio. And it started with death being the top outcome to be considered, second

pulmonary hemorrhage, third sepsis, then NEC, then ROP. Okay?

Ben Courchia, MD (20:10.19)
So, in terms of results, let's get into them now. So now we can understand a little bit what we're trying to do. They enrolled 116 out of 185 potentially eligible infants, for a recruitment rate of 63%. Of the 116 babies that were enrolled, 104 infants, or 90%, were randomized. In the randomized cohort, the mean gestational age was 24.3 weeks, with a standard deviation of

0.8 weeks. The birth weight was 714 grams. Breaking down that arm, 51 infants were allocated to the SMART arm with a mean gestational age of 24.4 weeks and a birth weight of 726 grams, and the other arm 24.2 weeks and 702 grams, so quite similar.

In terms of treatment characteristics of the 51 infants that were in the SMART arm, 45% received pharmacotherapy after their first echo, 17.6% after their second, and one patient after their third echocardiogram. The median age of pharmacotherapy initiation was two days, and

12 infants, or 24% of the cohort, never received pharmacotherapy at all. Of those receiving pharmacotherapy based on the SMART-PDA protocol, 25 received ibuprofen and eight acetaminophen as their primary pharmacotherapy.

Those are the 53 infants; 64% received PDA pharmacotherapy during their hospital stay at a median age of nine days, so after that initial seven-day period. And with regard to procedural closure, 9.8% in the SMART arm and 13% in the control arm underwent invasive PDA closure. So in terms of clinical outcome, they say that the Bayesian analysis showed that the SMART approach

Ben Courchia, MD (22:26.286)
was associated with a reduction in pulmonary hemorrhage, with a relative risk of 0.07 and a probability that the SMART approach is better than control of greater than 0.99. So let me just reframe that for you. It means that the Bayesian analysis showed that the SMART approach had a likelihood of reducing pulmonary hemorrhage

by 99%. And I'm just going to let that sit for a second, and go to the next one. It also sh—Yeah, okay, yeah, okay. Very astute observation, my friend. It also sh—

Daphna Yasova Barbeau (23:02.179)
 That's a lot of percents. Thank you, thank you. We've only been doing this journal club for five years, you know. I've got—

Ben Courchia, MD (23:12.404)
Uh-huh. It also showed a clinically meaningful probability of benefit with regard to NEC, with a relative risk of 0.39 and a probability that the SMART intervention is better than control of 0.92, which is about 92%. Which now makes you understand why

when that data came out, a lot of the centers were saying, I'm not staying, I'm not leaving my patient in the control arm, right? That doesn't make sense to me, because you're telling me that the likelihood of reduction is so much higher. So when we're talking about this, right? I think this is really quite

It was quite incredible. And that's also where that Bayesian analysis looks at the hierarchy, right? We're not talking about death, we're talking about pulmonary hemorrhage, which is your second outcome, and NEC, which is your fourth outcome in that hierarchy. Does that make sense?

Daphna Yasova Barbeau (24:31.991)
Got it.

Ben Courchia, MD (24:32.879)
Yeah. So for the other outcomes, the estimates were imprecise and centered around one. So I'm going to skip that. And with regard to pharmacotherapy-related outcomes in the SMART arm, 3.9% developed oliguria, and none developed GI bleeding within the initiation of treatment. So I'm going to leave this there for a second. The conclusions are that the trial—

the first to exclusively enroll infants born before 26 weeks' gestation, suggests that the sickest micropreemie with a high-volume PDA shunt identified through early comprehensive hemodynamic screening may benefit from pharmacotherapy. Future trials should examine the role of such selective pharmacotherapy strategies in the highest-risk infants, as specifically highlighted as a research priority by the American Academy of Pediatrics in the clinical report on PDA management.

So this is very interesting. And I think that, to me, what's very innovative is the approach of the Bayesian analysis, the win ratio that they did, and the outcomes we're discussing. So one of the things that, to me, is very

very peculiar is that we're not talking about the closure rates of these PDAs, right? That, to me, is the question lingering through the paper. It's like, well, how many of them got their PDA to close? And they purposefully don't; that's not important, right? That's not really what they're looking at. But it is something that people might say: well, did it work? Did it close the PDA? And so on and so forth. And I think this segues quite nicely to another paper that Souvik is

publishing, this time in JAMA Network Open, called Pharmacologic Therapies for PDA in Extremely Preterm Infants. And we're going to talk a little bit about that, where it's a bit of a comparison. But do you have any questions or thoughts on the first one, about the SMART-PDA trial?

Daphna Yasova Barbeau (26:34.873)
Well, phew, that was a big one. Well, I think, in our practice, this question about did it close the PDA? Right, maybe it doesn't close the PDA, but it has enough effect on the PDA that we still get a positive impact. Or it makes that PDA more restrictive, so it becomes less hemodynamically significant. But yeah, potentially the goal's not closure, right? And I think what's cool about this is

we know that exposure to the PDA is bad. But every time we try to prove that closure makes that exposure better, we've failed to do that. But I think what was lacking was a standardization of what's hemodynamically significant. And not all the studies were functioning off the same page of music, I think. So that's interesting. I wonder.

Ben Courchia, MD (27:27.289)
No, no. And the definition of a hemodynamically significant PDA, plus the implementation of those definitions into a randomized trial format. Yeah.

Daphna Yasova Barbeau (27:36.109)
Right, for sure, for sure. And I'm going to ask you, as a burgeoning expert in hemodynamics: what do you think will be some of the concerns about this study? What do you think the "we don't need a closed PDA" group will say about this study, including some of your colleagues in Montreal, right?

Ben Courchia, MD (27:59.463)
Well, I think they have strong opinions on that. But I think the argument here is that this is based off a Bayesian analysis, and the fact that in the end we're not talking about a tremendously large number of babies. And while I think this is going to come counter to the fact that

Daphna Yasova Barbeau (28:06.595)
That's right.

Ben Courchia, MD (28:29.894)
the PDA burden is individualized to each patient in that particular format. So yeah, I think it will be interesting. And again, we're talking about the first seven days of life, and we don't know if the PDA has been closed or what the effect of that was on the PDA itself. I think this is where people are going to say, I want to see the data bear out on the patients.

Daphna Yasova Barbeau (28:57.697)
All right, I can buy that. Carry on.

Ben Courchia, MD (28:59.216)
Yeah. So in JAMA Network Open, basically they prospectively compared the relative effectiveness of different PDA pharmacotherapy regimens in extremely preterm infants, those born before 29 weeks of gestation. So a little bit more mature population than the one we just spoke about, which was less than 26 weeks. So I'm going to try to go relatively quickly, because it's already been 30 minutes. But the

They conducted a multicenter, prospective, observational comparative effectiveness research trial in 19 tertiary centers in Canada. Infants born less than 29 weeks were included. They had to have a large left-to-right shunt at the PDA. Infants who received indomethacin in the first 24 hours of life for IVH were also included. And they had

four predefined interventions for the participants: standard-dose ibuprofen, adjustable-dose ibuprofen, indomethacin, and acetaminophen. These were the primary pharmacotherapies. The primary outcome was failure of primary pharmacotherapy, meaning needing additional medical or surgical intervention following an initial medication administration. The secondary outcomes included receipt of a repeat pharmacotherapy course, surgical closure, moderate to severe

BPD, necrotizing enterocolitis, IVH, sepsis, and so on and so forth. In this study, they had 1,356 infants between 2020 and 2023, mean gestational age 25.4 weeks, mean birth weight 828 grams. Of these patients, about 1,100 received PDA pharmacotherapy,

whereas 259 did not receive pharmacotherapy and were conservatively treated. Among the participating NICUs, eight used standard-dose ibuprofen, eight used adjustable-dose, three selected acetaminophen, and one selected indomethacin. And again, the decision on how to treat these infants was left at the discretion of the different centers.

Ben Courchia, MD (31:24.202)
So let me just get to the drug-dose effectiveness cohort. Okay, let me talk about the treatment effectiveness and primary outcome. So among the 1,100 infants who received PDA treatment, 42% experienced failure of the primary pharmacotherapy. In addition, 41.6% received a repeat pharmacotherapy course,

and 4.1% underwent interventional PDA closure. Importantly, there was no difference between the four regimens in terms of which one failed the most. So the comparison between all four treatments seems to be relatively the same. With regard to secondary outcomes, and using the standard-dose ibuprofen arm as the comparator, mortality odds were higher among infants who received acetaminophen,

with an adjusted odds ratio of 1.37. And mortality odds were lower among those receiving indomethacin, with an adjusted odds ratio of 0.4. Sepsis odds were also lower among those who received indomethacin, with an adjusted odds ratio of 0.21. Interventional PDA closure odds were lower among those who received adjustable-dose ibuprofen. And there was no difference in adverse effects, specifically post-treatment serum creatinine, bilirubin levels, and liver function tests

with any of the pharmacological interventions.

We already talked about that. When it comes to looking at other clinical outcomes for these infants who received PDA treatment, pre-discharge mortality was 14.5%, moderate to severe BPD 65.9%, NEC stage two 11.6%, and definite sepsis following initiation of treatment 22.8%. Then they compared the

Ben Courchia, MD (33:29.011)
kids who were treated as one big cohort versus the ones who were not treated. And here we have the treated group, which is about 1,100 infants, being compared to the 259 infants who were not treated. I'm sorry—right, did I say that correctly? The 259 infants who did not receive treatment. So for pre-discharge mortality,

the rate was 14.5% in the group that received treatment versus 15.4% in the group that was conservatively treated, meaning those that did not receive treatment. And that difference was not statistically significant. When we're looking at moderate to severe BPD, the rate was 55.2% in the treated group

versus 32.4% in the conservatively managed group. For NEC, the rate was 11.6% in the treated group and 6.2% in the conservatively managed group. And for definite sepsis, the rate was 22.8% in the treated group versus 21.2% in the non-treated group. When compared with

those who received pharmacological treatment, infants who received PDA therapy had higher odds of moderate to severe BPD, with an adjusted odds ratio of 1.91, and higher odds of NEC, with an adjusted odds ratio of 2.15. And the inverse—well, they did an inverse probability weighting, but I'll skip that for now. So I'm going to get to the sensitivity and subgroup analysis, and then I'll close out for today.

The per-protocol sensitivity analyses were largely consistent with the primary analysis. They demonstrated higher mortality odds with acetaminophen, with an adjusted odds ratio of 1.47. They also demonstrated, with indomethacin, lower mortality, with an adjusted odds ratio of 0.16, lower sepsis, and higher BPD odds.

Ben Courchia, MD (35:47.476)
Turning to the subgroup analysis, the notable findings include lower mortality odds with indomethacin within seven days of birth, with an adjusted odds ratio of 0.16, and higher BPD odds with indomethacin, both within seven days, with an odds ratio of 2.7, and after seven days. In addition, in the subgroup of infants born before 26 weeks—now we're talking about the same population as the SMART-PDA cohort that we were discussing—

indomethacin was associated with lower mortality and higher BPD odds. And in that same subgroup analysis by gestational age, adjustable-dose ibuprofen was associated with lower BPD odds in both gestational age subgroups. For infants below 26 weeks, the adjusted odds ratio was 0.33. And for infants at 26 weeks and above, the adjusted odds ratio was 0.38. The conclusions are that,

in this large multicenter comparative effectiveness research study, it really shows that the primary PDA pharmacotherapy failure rate was high. We're talking about 40%, and similar in extremely preterm infants, regardless of the medication that you choose. The use of acetaminophen seems to be associated with increased mortality odds, but this result should be interpreted with caution, again because of potential confounding by indication or contraindication.

Although no pharmacotherapy choice, as we described—i.e., ibuprofen, acetaminophen, or indomethacin—stood out as more effective. Overall, infants who were exposed to pharmacotherapy had higher odds of moderate to severe BPD and necrotizing enterocolitis, but lower mortality odds compared with infants who were conservatively managed. Future studies should approach—

and should investigate approaches for identifying extremely preterm infants at high risk for PDA-attributable morbidity, and infants most likely to respond to medical therapy. This is a great linkage to the other paper that we just reviewed. I'm going to want to stop here. I'm exhausted.

Daphna Yasova Barbeau (37:57.495)
I'm sure you are. Okay, so where does this leave us?

Ben Courchia, MD (38:01.203)
Well, I think it's very interesting, because now you see that this approach of just saying, hey, we use acetaminophen and then we treat everybody based exactly on what the PDA size is—it's not good. There's actually a great editorial in JAMA Network Open that I'm inviting everyone to read, authored by Marjorie McConey,

Daphna Yasova Barbeau (38:07.161)
Yeah, we treat everybody. Everybody is—

Daphna Yasova Barbeau (38:16.878)
Mm-hmm.

Ben Courchia, MD (38:30.078)
which basically makes this exact point: that the burden of a PDA is individualized. A 1.5-millimeter PDA means nothing to me unless I understand how it is affecting that particular baby. And there are some babies who are going to respond very differently to a 1.5-millimeter PDA, and their intervention should thus be different. And I think that when you go back to the intervention, right?

They're talking about including babies who basically have a PDA that's above 1.5 millimeters. That's the inclusion. And it highlights how, if your goal is to just measure the PDA in size and give a medication hoping that it will close, this is not the right approach, because number one, your closure rate is extremely poor, 40%. And the outcomes are not clear-cut. It's not like you give the medication and suddenly mortality, morbidity, everything goes down.

So I think that, to me, this reinforces what we just saw with the SMART-PDA trial, which is: I don't really need to know what the closure rate is. I don't need to know the diameter. I need to have a better assessment of which patient I'm talking about, and then tailor my treatment to that particular physiology. So I think this probably made things more complicated for everyone, but—sorry.

Souvik's email address will be in the description, and you can email him.

Daphna Yasova Barbeau (39:56.205)
That's right. That's fine. People can address that with him personally. No, I think this is helpful. I think it again just underscores this individual approach to medicine, right? We want to be able to give all the babies the same thing and have good outcomes. And that's

Ben Courchia, MD (40:02.311)
Yeah.

Daphna Yasova Barbeau (40:18.519)
I think that's what they're telling us. It's that we can't do it that way, especially for the PDA, but probably for a lot of the other things that we've tried to do in neonatology.

Ben Courchia, MD (40:22.834)
You can't, yeah.

Ben Courchia, MD (40:28.744)
Yeah, and to me, it's fascinating that the same outcomes we were talking about, specifically when it comes to NEC, really can swing one way or the other, depending on whether your approach is a bit shotgun-y, where it's like, I'm just going to look at the size and treat or not treat, versus a little bit more tailored. Whereas in the SMART-PDA trial, the data is sort of predicted, right? The SMART-PDA approach in the first seven days of life does have

significant implications for the reduction of that specific morbidity. So yeah.

Daphna Yasova Barbeau (41:03.885)
Yeah, for example, this reversal of flow in the aorta. Like, if the intestines aren't getting enough flow, it stands to reason that you'd be more likely to get NEC. So, okay. Yeah. You sure have.

Ben Courchia, MD (41:11.077)
Mm-hmm. Yeah, absolutely. We have to move forward, because I've taken way too much time. All right. I'll see you tomorrow. Bye.

Daphna Yasova Barbeau (41:23.191)
Okay.