#327 - 🫀 From the Heart - It's Complicated

Hello friends 👋

In this episode of From the Heart, Dr. Nim Goldshtrom and Dr. Adrianne Bischoff walk through a complex neonatal case involving sudden decompensation shortly after birth. With few early clues and no clear diagnosis, the team discusses how to approach circulatory shock, differentiate pulmonary hypertension from congenital heart disease, and manage critically ill neonates before imaging is available.

Using this case as a reverse journal club, the hosts break down relevant literature and decision-making pathways: when to start prostaglandin, when epinephrine makes sense even without low blood pressure, and why relying only on numbers like MAP can be misleading. They also examine the role of therapeutic hypothermia in unstable infants and the potential cardiovascular consequences of cooling.

Later, the conversation focuses on left ventricular dysfunction, balancing systemic and pulmonary circulation via the ductus, and using bedside markers like lactate and perfusion to guide treatment when echo isn’t immediately available. The episode closes with thoughts on autoregulation, cerebral protection, and the evolving role of emerging technologies in neonatal hemodynamics.

A real-world deep dive into diagnostic uncertainty, evolving physiology, and decision-making under pressure in the NICU.

----

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

Blood pressure:

Barrington KJ. Common hemodynamic problems in the neonate. Neonatology. 2013;103(4):335-40. doi: 10.1159/000349933. Epub 2013 May 31. PMID: 23736012.

Dempsey EM, Barrington KJ, Marlow N, O'Donnell CP, Miletin J, Naulaers G, Cheung PY, Corcoran D, Pons G, Stranak Z, Van Laere D; HIP Consortium. Management of hypotension in preterm infants (The HIP Trial): a randomised controlled trial of hypotension management in extremely low gestational age newborns. Neonatology. 2014;105(4):275-81. doi: 10.1159/000357553. Epub 2014 Feb 27. PMID: 24576799.

Dempsey EM. What Should We Do about Low Blood Pressure in Preterm Infants. Neonatology. 2017;111(4):402-407. doi: 10.1159/000460603. Epub 2017 May 25. PMID: 28538235.

Autoregulation and blood pressure in HIE:

Howlett, J. A., Northington, F. J., Gilmore, M. M., Tekes, A., Huisman, T. A. G. M., Parkinson, C., Chung, S. E., Jennings, J. M., Jamrogowicz, J. J., Larson, A. C., Lehmann, C. U., Jackson, E., Brady, K. M., Koehler, R. C., & Lee, J. K. (2013). Cerebrovascular autoregulation and neurologic injury in neonatal hypoxic-ischemic encephalopathy. Pediatric Research, 74(5), 525–535. https://doi.org/10.1038/pr.2013.132

Burton, V. J., Gerner, G., Cristofalo, E., Chung, S. en, Jennings, J. M., Parkinson, C., Koehler, R. C., Chavez-Valdez, R., Johnston, M. v., Northington, F. J., & Lee, J. K. (2015). A pilot cohort study of cerebral autoregulation and 2-year neurodevelopmental outcomes in neonates with hypoxic-ischemic encephalopathy who received therapeutic hypothermia. BMC Neurology, 15(1), 1–13. https://doi.org/10.1186/s12883-015-0464-4

da Costa CS, Czosnyka M, Smielewski P, Austin T. Optimal Mean Arterial Blood Pressure in Extremely Preterm Infants within the First 24 Hours of Life. J Pediatr. 2018 Dec;203:242-248. doi: 10.1016/j.jpeds.2018.07.096. Epub 2018 Sep 20. PMID: 30243537. 

----

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

Dr. Adrianne Bischoff: Hi everyone. Today on From the Heart: It's Complicated, we are tackling a challenging neonatal cardiac case. Nim and I thought this would be a great opportunity to walk through our thought process, break down the physiology, discuss key decision points, and explore some of the nuances of managing critically ill neonates. It's kind of like doing a journal club—but backwards. We're hoping that through our discussion of these cases which we'll do from time to time, we can bring in relevant literature so it's not just what Nim and I think, but grounded in science. We want to explore why we manage things the way we do and how we think about them. Is that right, Nim?

Dr. Nim Goldshtrom: Yeah—evidence-based medicine in practice, right? Reading an article and applying it is great, but what a great way to show everyone how we think and work—how we apply evidence in practice and go backwards from a case toward that evidence.

Dr. Adrianne Bischoff: Cool. So let's dive right in.

Today's case involves a 39+1-week infant born by spontaneous vaginal delivery. The baby was vigorous at birth, though there was a history of thin meconium-stained fluid. Apgar scores were 8 at both one and five minutes. The baby was placed on the mother’s chest. About seven minutes of life, was noted to be dusky. The team brought the baby to the warmer, placed a sat probe, and saw saturations of 30% on room air. PPV with oxygen supplementation was initiated, but there was no significant improvement. The baby was moved to the NICU, intubated at about 30 minutes of life, given a normal saline bolus, and a transfer call was initiated to a level III unit.

By that time, the baby had already received surfactant for unclear reasons. During the transfer call, the baby was on mechanical ventilation: volume control at 7 mL/kg, PEEP of 6, rate of 30, and FiO2 at 90%. The most recent arterial gas showed a pH of 6.9, PCO2 of 67, PO2 of 27, and a base deficit of 20. Vitals at that time: heart rate 160, blood pressure (via UAC) 69/56, and lactate 10.2.

So Nim, hearing all this—what are some of your key concerns at this stage?

Dr. Nim Goldshtrom: I’m just happy this baby is alive and doing relatively well. But yes, there’s severe acidosis on a vent, with a lactate of 10. That’s shock. The question is what type—maybe distributive, but more likely cardiogenic or mixed shock with hypoxic acidosis. Pulmonary hypertension comes to mind first. Congenital heart disease is another. Some sort of failure of transition. Maybe poor function due to sepsis or cardiogenic causes.

The next steps would be: get pre- and post-ductal sats, four-limb blood pressures if possible, check for saturation differentials, assess signs of cardiac output and systemic oxygen delivery, perfusion signs, lung exam, x-ray findings, and any other supportive data we can gather.

Dr. Adrianne Bischoff: So I have limited information because this was during the transport call. They didn’t have four-limb BPs—too much going on and too much concern about the baby's condition. They had a UAC for BP readings. Pre- and post-ductal saturations after intubation: pre-ductal was in the 70s, post-ductal in the mid-50s. The baby appeared shocky—pulses were okay, but the baby looked mottled with poor perfusion. It was too early to assess urinary output, but clear signs of poor systemic perfusion. X-ray wasn't very informative—no cardiomegaly, lung fields were okay, maybe a bit TTN-ish. They didn’t have a pre-surfactant x-ray. We could debate the rationale for giving surfactant to a term baby with this presentation, but we don’t have enough information for that.

So we talked about the differential: shock, sepsis, PPHN, congenital heart disease.

The first article I want to bring up is about distinguishing in our minds the difference between hypoxemia of cardiac origin (i.e., CHD) and pulmonary hypertension with normal cardiac anatomy. The first article is not new, it’s from 2010 from the Toronto group. They studied hypoxemic neonates during transport, trying to determine the accuracy of clinical diagnoses between CHD and PPHN.

They classified babies based on clinical impression—suspected CHD, suspected PPHN, or questionable—based on history, physical exam, labs, chest x-ray, and initial response to treatment.

Their table is a great reference for trainees. Some highlights:

• CHD: dysmorphic features, murmur, weak pulses, significant right arm–leg BP difference, saturation not improving with oxygen weaning, cardiomegaly on CXR.

• PPHN: fetal distress, low Apgar, poor pH, labile when handled, worsening sats when oxygen is weaned, abnormal lung parenchyma (PPHN is a vascular disease, but the majority of neonates with PPHN will have lung disease as well).

They were actually about 88% accurate in their clinical predictions, which I thought was impressive. So I’ll put it back on you, Nim. Based on what I’ve told you so far, what do you think? Which category do you lean toward?

Dr. Nim Goldshtrom: Just based on what we have so far, it’s hard to say. There’s bad acidosis, high lactate. In my practice, I always worry about ductal-dependent lesions, but most don’t present this acutely. Only a handful of congenital cardiac lesions will collapse in the first 30 minutes of life if not handled appropriately. Those include obstructed TAPVR (which could look similar to this case), transposition with poor mixing (but the baby had an FiO2 of 90%, and we don’t know if they ever had normal sats), and HLHS with a restrictive atrial septum (but this doesn’t feel like a classic case). Honestly, I don’t have a solid guess yet—this could still be one of those three CHD types or PPHN. Let’s get more data.

Dr. Adrianne Bischoff: Yeah, I agree. The baby fits a bit on both sides of that diagnostic table. The team felt the same—they weren’t confident enough to say “This is definitely PPHN, let's give nitric,” but also didn’t feel it fully fit with CHD. So, in the absence of certainty, they started prostaglandin and epinephrine due to the shock picture—even though BP was technically okay. And I think that was appropriate. You don’t need to wait for hypotension to start cardiovascular support. This baby had signs of poor cardiac output. They started prostaglandin at 0.03 to maintain ductal patency and epinephrine at 0.05 for presumed cardiac dysfunction. Due to the possibility of CHD, they held off on starting nitric oxide during transport.

Upon arrival at the NICU, now nearly 4 hours of life, FiO2 had been weaned to 60%. Pre-ductal sats were in the mid-to-high 90s; post-ductal sats were in the low 80s. Lactate had improved to 3.3 (was 10.6 initially). The x-ray showed mild venous congestion, but nothing dramatic. Given the history of an apparent asphyxial event and hypoxia with acidosis, the baby had an abnormal Sarnat exam consistent with moderate encephalopathy and qualified for therapeutic hypothermia.

Now, with all this information—would you classify this baby more clearly into one diagnostic category? And would you consider starting inhaled nitric oxide?

Dr. Nim Goldshtrom: Great questions—and I do have a thought, though it might not fall neatly into either category. But first, I want to underscore an important point you made. One is not waiting for “hard” markers to start shock medications is critical. This baby had a MAP probably in the low 60s. But the nuance there is that you have a pH of 6.9, lactate of 10, and a narrow pulse pressure with a “normal” blood pressure. That’s compensated shock by definition. You don’t wait and assume that the baby is going to improve. You act. I completely agree with starting epinephrine for cardiogenic shock. This kid, with a narrow pulse pressure, is close to uncompensated shock. Trying to maintain trying to maintain SVR high in the face of almost certainly low cardiac output, they need beta and inotropy rapidly. It's very weird because you have a heart rate of 160, you have a normal to high blood pressure that's narrow, and a lactate of 10. It can look like it's not as dangerous as it is, but this is about to collapse and arrest physiology. There is no more data you should need to try to recognize compensated shock, which is just holding onto your vitals before you end up collapsing.

Dr. Adrianne Bischoff: Exactly. And I just want to emphasize we don’t keep escalating epinephrine unless we have to. We're aiming for inotropy here, not more vasoconstriction. Epinephrine effect’s are dose-dependent. At low doses, like which was what was started here, the main action will be on inotropy, which is what we were aiming for. But if you keep escalating, especially once you're getting to 0.1 and more, then you're gonna start having a lot of alpha-adrenergic effects, and you can increase the SVR, which in a kid that has evidence that they have strong SVR, you might actually make things worse. Remember, we're not treating the number, but we're treating the patient. We have evidence that using this dose, once the baby arrived, that they are improving, irrespective of what the blood pressure was. You can see that I didn't even mention what the blood pressure was at this point, because it's irrelevant. The lactate is better, the saturations are better, the perfusion is better. So I just wanted to make that point about the dosage of the medication.

Dr. Nim Goldshtrom: I'm sure I'm speaking to everyone who knows this, but you're just going to be watching this baby very closely and doing very frequent assessments of the exam, lactates, and gases. And if available, tools like near-infrared spectroscopy (NIRS) can help us monitor regional perfusion, especially lower body perfusion (either renal or abdomen) as a trend for the venous oxygenation system to complete the VO2-DO2 circuit. And for us, we do find it helpful in picking up early signs of shock and then the recovery of shock when you're probably resuscitate if you use those in your markers.

And so as you mentioned, you've got the patient four hours later on a lower FIO2 now with a pre and post-ductal gradient, but having high sats pre-ductally, low sats post-ductally, and venous congestion. This suggests an issue with arch flow. It’s likely that unoxygenated blood is being shunted through the PDA to the descending aorta—hence the lower post-ductal sats. So if you can't send enough blood out of the left side of your heart, where the duct and the right side of the heart have to contribute to systemic perfusion, and your pulmonary vessels are congested, to me, that story fits with your left side has failed for some reason, and the right side is compensating.

This almost sounds like a coarctation-type physiology. If we had four-limb pressures, that might clarify it further.

Dr. Adrianne Bischoff: That’s a great point. I think a lot of people might jump to the most common explanation: an asphyxiated baby needing lots of oxygen. Venous congestion could easily be mistaken as TTN, RDS, or meconium aspiration. They might assume it's a right-sided problem and consider pulmonary vasodilators—especially with the venous congestion on x-ray.

Dr. Nim Goldshtrom: Yes, that’s probably the most common interpretation—and not wrong, especially without more imaging. It would be reasonable to think common things are common, and a failed transition, with wet lungs, and this looks like PPHN physiology. It’s not uncommon for infants with horrible cord gases and PPHN have lactates of 10.

But the key detail here is that epinephrine worked. The treatment for PPHN is not epi. If this were pure PPHN, we’d expect nitric oxide to help, not epi. That response makes me think there’s a left-sided issue—or at least a mixed picture. The team tried something that is not a direct pulmonary hypertension medication, and saw clinical improvement. If epi was on but the baby did not improve, then I would’ve gone a different direction. It could be PPHN with cardiac stunning, leading to cardiogenic shock. But the baby responded to inotropy, which pushes me to consider that this was shock, maybe mixed shock. It’s confusing, but fascinating. I’m curious to see where this case goes next.

Dr. Adrianne Bischoff: Yeah. And I think along those lines, the more traditional train of thought would be that this is an asphyxiated baby we're going to start cooling. So first, we would consider the most common or natural pathway—pulmonary hypertension and heart dysfunction—most commonly right heart dysfunction in the setting of pulmonary hypertension, but also in HIE itself, right? We have Dr. Giesinger's work, both in The Blue Journal and subsequently in Archives, where she evaluated a large prospective cohort of patients with HIE undergoing cooling. Right heart dysfunction hadn't really been documented as a major feature of HIE until she did that work—especially with the level of detail and the quantitative measurements of RV dysfunction. Not only was it the predominant phenotype seen in patients with HIE, but having right heart dysfunction was significantly associated with worse outcomes—both in the short term, with higher mortality and worse neurological outcomes including poor brain MRIs, and in the long term with poorer neurodevelopmental outcomes in survivors.

So the typical thinking would have been: this is pulmonary hypertension and RV dysfunction. And maybe people could argue, “Well, you gave epi; maybe you improved the RV.” But I would counter that by saying RV dysfunction in HIE patients, if you don't treat the afterload problem—which is the pulmonary vasculature—typically doesn't improve this quickly, certainly not in two hours, and especially not without a pulmonary vasodilator. Just giving epi to what frequently are non-contractile RVs that we see on echo in these asphyxiated babies does not usually result in significant improvements in saturations, perfusion, lactates, and so on. Is that your experience as well, Nim?

Dr. Nim Goldshtrom: Yeah, absolutely. And this is something we both see in different ways. Clinical exposure and experience give us insights into how certain things recover over time. For example, pH crises after cardiac surgery have a desaturation arc that's very different from a respiratory desaturation or a pure right-to-left cardiac shunt. You get used to observing it over time and differentiating them. So I agree—this is a different recovery arc than you'd expect in a baby whose only issue is dysfunction.

Dr. Adrianne Bischoff: And along the same lines—since we're talking about HIE and cooling—what’s your experience, or what do you typically see in terms of heart function when you start cooling? Do you notice anything that suggests cooling might make cardiovascular function worse?

Dr. Nim Goldshtrom: Absolutely. The literature is rife with those papers showing both LV and biventricular dysfunction during cooling. This isn't evidenced-based research, but in clinical practice, many of our kids need to be started on milrinone, epi, or dobutamine depending on which ventricle is struggling during cooling to maintain a decent cardiac output. We're not just dealing with a specific type of hypotension, as cooling also induces vasoconstrictive states. We're trying to support LV or RV recovery, so most of them go on milrinone. Some people might be hesitant because the patient is normotensive, but if they're still showing signs of cardiac dysfunction and delayed lactate clearance, you can tolerate the afterload effects of milrinone while allowing for lusitropy and some gentle inotropy. But yes, cardiac dysfunction during cooling is a well-described phenomenon.

Dr. Adrianne Bischoff: Yeah. Your colleague Angelica Vasquez—who was a trainee here—did a small study during her fellowship at the University of Iowa in 2024. She published it in the Journal of Perinatology, and we'll link it in the show notes. She looked at 10 babies who had targeted neonatal echoes before and after the initiation of cooling, without any other changes in therapy during that short interval.

What she found was that post-cooling was associated with worsening markers of pulmonary hypertension and increased PVR, with a decrease in cardiac output. There was also a decrease in heart function, though that wasn’t quite as significant in this small cohort.

That’s another important point. While it’s not the central discussion here, in really sick babies, cooling might actually make things worse. That’s a bit controversial, but I personally think that sometimes we rush into cooling too quickly because we want to protect the brain—which, of course, is critical. But if the baby decompensates rapidly, ends up on 100% oxygen, barely maintaining themselves, and then we start cooling... now their sats are worse, they might require ECMO or die from a pH crisis, that’s not helping either. So I tend to slow down cooling a bit—waiting for at least some degree of cardiorespiratory stability before starting, if possible. That makes the process smoother and decreases the risk of decompensation. We can get into discussions about cooling while on ECMO and all that, which may be too much for today, but I’m happy to hear your thoughts.

Dr. Nim Goldshtrom: No, it's true. Cooling is an escalation in critical care. The body, under hypothermic conditions, is not going to function better. The liver won't be happy. The heart won't be happy. It's a semi-procedural process. You’re actively bringing the body to a target temperature of 33°C. It’s not a simple “flip the switch” process. You need to make sure you have cardiac and respiratory control, matching of DO₂ and VO₂, and good sedation so the patient isn’t distressed and increasing demand. This is just as important as the decision to cool or not.

Dr. Adrianne Bischoff: Exactly. When I run transport calls, I usually recommend that outside providers—though not my transport team—maintain normothermia in most patients. That’s for two reasons. First, many outside providers aren't used to cooling. They just turn off the warmer, and if the baby is truly asphyxiated, the temperature tanks. Unless they can really monitor and control the temperature, I prefer they wait for my team. Second, from a stability standpoint, if cooling is initiated, PVR might rise and they might not have resources like nitric oxide or the expertise for the necessary escalation in cardiorespiratory support. So I usually advise them to hold off.

Dr. Nim Goldshtrom: Absolutely. You have to know your system. Cooling on transport is possible, but you have to be confident that the baby will be supported before you arrive—and that your team can continue that support effectively throughout transport.

Dr. Adrianne Bischoff: All right, back to this case. The question on everyone’s mind: what did the echo show? What’s the diagnosis? You were right, Nim. This baby had severe left ventricular systolic dysfunction, with an ejection fraction of 38%. For those less familiar with echo numbers, normal in a neonate would be more than 55–60%. The baby had low cardiac output, with a left ventricular output of 65. In cooling babies, I sometimes see lower-than-expected LVOP, but 65 is definitely on the low side—90s would be more typical.

The baby also had mild right ventricular dysfunction, but the LV dysfunction was more pronounced. There was an unobstructed aortic arch, no structural heart disease, and normal pulmonary vein drainage. The coronaries were fine. At this point, the baby was on epinephrine at 0.05 and prostaglandin at 0.03. There was a large PDA, five millimeters, shunting bidirectionally. So now that you know the diagnosis, how do we approach severe LV systolic dysfunction with low cardiac output in a baby on these medications? What are your thoughts?

Dr. Nim Goldshtrom: You know, feed and grow, just feed and grow. No, we kid, obviously. Just to reiterate some of the points you're bringing up with the echo and the clinical picture...if we're presenting a case with spot numbers, but even with the description of pre- and post-ductal sats and congestion and my running theory of left-sided dysfunction, almost certainly this kid was desaturating during periods of time in the unit. And this almost certainly would have looked like PPHN. Even I might have started nitric. The caution I have in these cases is that echoes are super helpful, but anytime I see an X-ray and the story is of a kid who was fine, fine, fine—and then very quickly fell apart after being on mom's chest—my fear is always, just like you said, PH or CHD. When the diagnosis could potentially be obstructed veins or PH, and my X-ray is weird, I'm always just hesitant to try PH therapies first, because they could make it worse. That’s when I run to the function algorithm and say, “Let’s just support the heart until we can get an echo,” because that’s what I’m definitely afraid of. It’s such a problem that two diseases can look exactly the same, and one therapy can fix one and make the other catastrophically worse. And obviously, not everyone has immediate access to an echo. This pilot had to be moved to a center to get it—and that’s sadly the fate for a lot of obstructed TAPVRs who are born in low-risk centers and then have to be rapidly escalated before a diagnosis is found. And everyone’s stuck in this place, right? Do you start nitric? Do you not?

And to that point, I just want to go back to the very first things: normal heart rate, narrow pulse pressure, normal blood pressure, and high lactate—start inotropic support. That is shock by definition. Whether you're debating, “Well, there's desats and it could be nitric plus shock,” yes. And there's a good chance that in our center, if this kid was born like this and we couldn't get an echo right away, we would have done everything: Epi and nitric and waited for the echo. And now that you're displaying an open duct with bidirectional shunting and severe LV dysfunction as the likely physiology, this will take time. Part of the question is, why did the kid go into shock? What's happening anatomically? I’d want to know, again, that you have no structural heart disease, which is great. You’re describing normal veins, normal coronaries, no coarctation, arches are not obstructed or hypoplastic. You don’t have a Shone’s-like picture with small mitral valves or small outflow structures. It really is just a normal heart that’s in shock. This is going to take time.

There’s a little bit of literature on this—what’s the right drug of choice? Dobutamine, milrinone, epinephrine? I’m not up to date on the most important studies about which one has the best efficacy for heart recovery or fast architecture, but the most important thing is that they give you options. They offer options for normotension LV dysfunction and for hypotension with normal function. I have hypotension and need beta and alpha? Great, that’s epi. I have normotension but the function’s really bad and I need to get this heart pumping? Dobutamine. I have dysfunction with normal tension and it’s both LV and maybe RV? Then maybe milrinone could help. That’s my broad thinking. How do you take these drug classes and use them in your armamentarium? And is there evidence that drives your choice between one or another for these kinds of conditions?

Dr. Adrianne Bischoff: Yeah. Between dobutamine and epi, my clinical practice is similar to yours—it’s dependent on the blood pressure and the degree of dysfunction. For mild to moderate heart dysfunction—whether LV or RV—when the blood pressure is normal or just slightly low, I think dobutamine is fine. It’s a cleaner drug; it causes less confusion, especially around hyperglycemia or high lactate, which can sometimes come as side effects from epinephrine (not necessarily related to the underlying dysfunction).

For severe dysfunction, or for hypotensive babies, I tend to go with epi—just because you get more bang for your buck. But I don’t think either is wrong. If you only have one or are more comfortable using one, I think that’s fine.

I try not to use dobutamine at more than 5–10 mcg/kg/min. If I’m at 10 and not seeing improvement in heart function or perfusion markers, I tend to switch to epi. I feel that escalating past 10 on dobutamine gives too much chronotropy without much added cardiac output or stroke volume.

With epi, again—it's dose dependent. If I’m using it for inotropy, I try not to go above 0.1 mcg/kg/min, because I don’t want to worsen things from the alpha standpoint.

But I’ll add a little caveat on milrinone—my project, so I’m going to toot my own horn here, probably not kosher, but I’ll do it anyway. We looked at babies with hypoxemic respiratory failure and pulmonary hypertension who received milrinone as adjunctive therapy on top of nitric. We compared cooling babies and non-cooling babies. In the cooling group, a very small cohort, we had a lot of trouble. Some babies became severely hypotensive with milrinone. We have to remember milrinone is renally excreted. Babies with HIE are prone to AKI, and they often aren’t peeing much due to SIADH. Who knows what therapeutic hypothermia does to drug clearance? But we had a few babies with catastrophic hypotension—single-digit blood pressures—who needed significant escalation in vasopressors to recover. They all survived, but had worse outcomes and higher inotropic scores. So here at Iowa, we’re quite cautious about using milrinone during therapeutic hypothermia. We usually don’t.

That said, I know of many centers who do use it. Our sample size was small. It’s possible that effect only occurs in babies with severe AKI or borderline BP. I don’t have the nuance or expertise for that. But at Iowa, if we need inotropes in cooling HIE babies, we use dobutamine or epi. We reserve milrinone for post-cooling.

Okay, what if things aren’t improving? Do you continue to escalate or add a second agent?

Dr. Nim Goldshtrom: Again, great points that I’ll add to. Milrinone has pros and cons. Continued hypotension suggests other parts of the system aren’t working. The cardiovascular system is interesting and built in wonderful ways: preload, contractility, afterload. You’re fixing some things with inotropes. Volume shifts typically aren’t severe, so extra fluid resuscitation isn’t common. But a neonate who is mildly sedated or recovering from shock probably has vasoplegia—not uncommon in distributive or multi-compartment cardiogenic shock. This is exactly how some kids present post-congenital heart surgery—volume and inotropes first. When they’re still hypotensive, we assume refractory vasoplegia and start pressors. Vasopressin is common, if you need more alpha. If you're already on epi, you’re getting some alpha—but not direct alpha like with other drugs. You could use norepi and epi, or much harder drugs like phenylephrine if needed.

If echo shows improving contractility, and no collapse suggesting low preload, and the child isn’t volume responsive, then yes—assume vasoplegia and add an afterload-increaser, like low-dose vasopressin or norepi or phenylephrine to carry them through the inflammatory phase until it has resolved.

Dr. Adrianne Bischoff: Can you educate us a little more, Nim, on BP targets and autoregulation in these patients?

Dr. Nim Goldshtrom: Oh boy. This is Keith Barrington’s world—he’s spent a career on this and we still don’t have a perfect answer. But we’re getting closer with new technologies.

Back in the 2010s, groups at Texas and Hopkins looked at autoregulation. It’s now a tool we can use to guide BP responses in the brain. Jennifer Lee’s group looked at therapeutic hypothermia using NIRS and BP to evaluate optimal BP during and after cooling, and neurodevelopmental outcomes. In term neonates around 40 weeks, optimal BP was about 51 ± 5. That’s close to what many of us use: gestational age = MAP. The deviation is: do we aim for the 10th percentile or 50^th for best outcomes?

In preterm infants 24-28 weeks, studies show the safe autoregulatory range is actually about 10 mmHg above GA in the first few days of life.

Dr. Lee’s group saw that time spent above the lower limit of autoregulation was associated with less white matter injury on MRI. They didn’t intervene—they just tracked BP vs. NIRS-derived autoregulation zones. This suggests the brain wants more BP than what we assume based on GA alone. But we don’t yet have evidence that treating to that higher number improves outcomes—or doesn’t cause harm. I gently aim slightly above GA-based MAPs in term infants, especially if they're already on inotropes, but I’m cautious. I don’t want therapeutic creep without proof of benefit.

Dr. Adrianne Bischoff: Yeah, all great points. And along those lines—how much do we really need to “fix” the problem? Dr. Giesinger’s group shows RV dysfunction is associated with worse outcomes. But is that causal, or just a marker of injury severity? If we improve cardiac output, do we improve brain outcomes? Maybe. But we also have to worry about reperfusion injury. We don’t want to go from low cardiac output to supranormal, which might make injury worse. So again—it’s about balance, and we need more physiological data to guide us.

Dr. Nim Goldshtrom: That’s the future. We’ve got the tools. The challenge is integration—and powering multicenter studies. Hemodynamics and autoregulation tools aren’t available everywhere. Even when it is, coordinating centers to gather data efficiently is hard. But this is where AI can help—linking big datasets, removing barriers of time, geography, and systems. We just need the collaboration.

Dr. Adrianne Bischoff: 100%. So, back to our case. We’ve got an asphyxiated baby with unexpected LV dysfunction.

Core priorities in managing this baby should include:

1. Support myocardial function — optimize inotropy. Use epi or dobutamine. I wouldn’t use milrinone during cooling.

2. Minimize myocardial oxygen demand — already cooling and intubated, but control agitation or pain.

3. Maintain ductal patency — in a neonate with severe LV dysfunction where the RV was actually not that bad, there is a role for maintaining ductal patency because functionally this works like a hypoplastic left heart, where you actually want right to left shunt in order to support the systemic circulation. So I don't want this baby to sat 100%, because at that point in time I might be switching my ductal shunt to be more bidirectional or to be more left to right, and I'm not going to be supporting flow. I would prefer to have blue flow than to have pink blood with no flow, which ends up being a gray baby. I would always rather have a blue baby than a gray baby.

4. Balance systemic/pulmonary flow — along the same lines, we want to make sure we are balancing the pulmonary and the systemic circulation. So we do things that will help promote right to left shunting, including providing adequate PEEP, targeting slightly higher CO2’s, and targeting lower oxygenation (depending on the severity of the situation) until the LV starts recovering. As the days go by, if your lactate improves and your pre/post-ductal saturation split is decreasing, that might be a clue that the LV is recovering, and the circulation is no longer as dependent on that right to left flow.

5. Sequential echoes — this will help guide when to stop PGE and titrate down epi.

Dr. Nim Goldshtrom: Great summation. One more thing—time is your friend and your enemy. Physiology will change. Ductal gradients will disappear. When sats are high 90s pre- and post-, you might not even need an echo to know it's time to wean PGE. You've offloaded the LV, the function is better. So use this dynamic picture and use the clinical tools. Like we mentioned, heart rate + blood pressure + lactate = shock, start treatment. Just reevaluate often. Let the baby evolve. That’s a huge part of managing this physiology.

Dr. Adrianne Bischoff: Yeah, absolutely. These are the cases that keep me up at night—but they also push us to be better clinicians.

Dr. Nim Goldshtrom: Literally and metaphysically—keep us up at night.

Dr. Adrianne Bischoff: So for the audience—if you enjoyed this discussion and format, let us know! Send us your thoughts, cases, or questions. You might hear them in a future episode of From the Heart.

Dr. Nim Goldshtrom: Absolutely. Thanks for joining us. We loved it, and we hope you did too. See you next time for more real cases, real challenges, and real-time learning in neonatal hemodynamics. Take care