#450 - [Journal Club] - 🫀 From The Heart - Is Dopamine Still Defensible as First-Line for Neonatal Septic Shock?

In this double-blind randomized controlled trial, Adrianne and Nim examine whether norepinephrine outperforms dopamine as a first-line vasoactive agent in neonates with fluid-refractory septic shock. The primary outcome, shock reversal at 30 minutes, was not significantly different between groups, at 32 percent for norepinephrine and 46 percent for dopamine. Secondary outcomes including mortality, IVH, NEC, and need for additional vasoactive support were also similar. The episode critically examines the methodological limitations of the study, including unclear sepsis definitions, absence of echo phenotyping, and unusually high starting doses, and asks whether the field needs better tools before these questions can be properly answered.

----

Norepinephrine versus Dopamine for Septic Shock in Neonates: A Randomized Controlled Trial.Mazhari MYA, Priyadarshi M, Singh P, Chaurasia S, Basu S.J Pediatr. 2025 Jul;282:114599. doi: 10.1016/j.jpeds.2025.114599. Epub 2025 Apr 17.PMID: 40252959 Clinical Trial.

Transcript

Nim Goldshtrom (00:00.000)
We have covered sepsis physiology and we have covered steroids. The only tool left to discuss is which drug to use during septic shock.

Adrianne Rahde Bischoff (00:44.356)
Yeah, so I am going to go straight into a controversial and hot topic. We are going to talk about a study that at the time of this recording has not officially been published yet, as it is coming out in the July edition of the Journal of Pediatrics. It is Norepinephrine versus Dopamine for Septic Shock in Neonates, also a randomized controlled trial. The first author, and I hope I am pronouncing this correctly, is Mohammed Yusuf Ali Masari, and the senior author is Shriparna Basou.

As we have touched on across the last two articles, a lot of practices for fluid-refractory shock are extrapolated from adult and pediatric guidelines, with recent guidelines favoring norepinephrine over dopamine as a first-line agent based on evidence showing benefit in all-cause mortality and a better cardiovascular profile in adults. In the NICU, historically, dopamine has been the most commonly used agent. Some of the concerns around dopamine include that it exerts its action through release of norepinephrine from sympathetic nerve endings, and neonates may have limited reserves of that. Additionally, the metabolism and clearance of dopamine is dependent on postmenstrual age. Finally, norepinephrine may have a more favorable pulmonary vascular resistance profile compared to dopamine, which may be beneficial in cases of pulmonary hypertension commonly seen in sepsis, though we just discussed from the Toronto paper that this may not always be the case.

This was a double-blind parallel-group RCT enrolling both term and preterm neonates with fluid-refractory septic shock. Septic shock was defined as blood pressure below the third percentile for gestational age plus two additional features, which could include low urinary output, prolonged capillary refill time, elevated lactate or metabolic acidosis, or low SVC flow. Fluid refractoriness was defined as persistence of shock after one or more normal saline boluses of 10 cc per kilo, and the diagnosis of sepsis required either a positive blood culture or a positive sepsis screen with compatible clinical course, which we can discuss.

The primary outcome was shock reversal at 30 minutes, defined as correction of hypotension, normalization of capillary refill time, and normal pulses with warm extremities. Per protocol, norepinephrine was started at 0.2 micrograms per kilo per minute and dopamine at 10 micrograms per kilo per minute. If shock had not reversed at 15 minutes, norepinephrine was increased to 0.3 and dopamine to 15. If shock still had not reversed, other drugs could be added at attending discretion. Crossover was generally not permitted but could occur if infants were refractory to multiple vasoactive agents.

They enrolled 80 babies, 41 in the norepinephrine group and 39 in the dopamine group, with three crossovers in the norepinephrine group and five in the dopamine group due to refractory shock on at least three vasoactive drugs. Forty-nine percent of babies had culture-positive sepsis, which raises my first question: were the others actual sepsis or a different type of shock? I was not entirely clear on what sepsis screen positive meant here. Was it based on CRP, left shift, leukocytosis? That was not well defined.

I also found it a bit odd that the median age at enrollment was four days, with an interquartile range going from roughly 28 to 48 hours up to seven to eight days of life, which would suggest most were early onset sepsis. But then they report 65 percent late onset sepsis in each group, which does not quite add up. I am not sure where that discrepancy comes from.

Either way, the primary outcome was not different. Shock reversal was present in 32 percent of the norepinephrine group and 46 percent of the dopamine group. The need for additional vasoactive agents and hydrocortisone was similar between groups. Other outcomes including shock reversal at any time point, time to shock reversal, hyperglycemia, severe IVH, ROP, NEC, and all-cause mortality were all similar as well.

NIRS values were lower in the dopamine group after 24 hours, and those babies tended to be more acidotic, but other physiological variables were similar between groups. There is no description of underlying physiology, echo data, or clinical phenotyping. Was this vasodilatory shock? Cold shock? Myocardial dysfunction? Were there shunts, and in which direction? What were the pulmonary hemodynamics? Was there hypoxemic respiratory failure? This is also a mixed population of term and preterm infants, and the definitions remain unclear. What does sepsis screen positive mean? How was SVC flow assessed, and what is the reliability of that measurement? If they did echo to assess SVC flow as part of their shock definition, why were other echo parameters not reported? SVC flow assessment in the echo world is not highly standardized and has significant methodological limitations.

I commend the group for conducting an RCT on two very important drugs, but there are significant methodological problems. I do not think this will change my practice in any shape or form. I need more information. The starting doses also surprised me. I would not start a baby at 0.2 micrograms per kilo per minute of norepinephrine. That is an extremely high dose. And even at those high doses, shock was not reversing. I am not sure what to make of that.

Nim Goldshtrom (07:55.844)
Go big or go home. That is just their philosophy. I have to say, if I am at 0.2 of norepinephrine, I am calling ECMO. That is a dose.

Adrianne Rahde Bischoff (08:10.100)
Exactly. So what are your thoughts, Nim?

Nim Goldshtrom (08:13.100)
You know, I think in trying to focus on TNE, septic shock physiology, and the options available, we are showing just how hard this work is. We have to commend these authors for trying. As people who write and publish our own research, we know it is difficult and we get challenged on a lot of things. It is not to say you cannot or should not pursue these questions, but it is genuinely hard to do good quality work with limited subjects and complex exposures.

That said, all the planning in the world will still produce problems, and the lesson is to think four times as carefully about your methods before you pursue your study, because that is where the answers live. These papers are giving us indications, yes, maybe no difference between the two drugs, but dopamine continues to show time and again that it affects multiple organ systems in ways we do not fully control. From the pulmonary system to the cerebral system to the heart, it is a drug with unreliable receptor effects when we now have other choices and tools like TNE that allow targeted approaches focused on much narrower and better-defined receptor classes.

Adrianne Rahde Bischoff (09:41.126)
And objective data rather than subjective attending impression. Whether the cap refill has improved, whether the extremities are warm. I could be swayed one way or another depending on which drug I am more comfortable with.

Nim Goldshtrom (09:57.060)
I know. And it is funny, because I am sure you do this too. On rounds when we are teaching residents and clinical staff, we always point back to the clinical exam and clinical picture. And yet when we discuss it from a research standpoint, we have to admit it is not reliable, right? Not without good interoperator reliability, and even with the same person, the exam changes over time. A lot of data points, and I harp on NIRS here, can be objective numbers but still limited. With NIRS it is really the direction and trend over time that matters most, not a single value. You have to know how to use your data point in context.

I am hoping the future in this data-centric world makes these studies better in profound ways. Take the hydrocortisone placebo paper. They had Phillips monitors, likely in a large regional hospital in India, pulling continuous data points, and they had a randomized trial and a system in place. It would have been interesting to think about secondary metrics that capture compensated shock beyond heart rate, blood pressure, and SpO2. Things like variance and skewness in data variability that can suggest compensated versus uncompensated shock. Integrating vital signs with TNE data in real time could be the future of how we design these studies. It is a huge lift and requires a completely different way of thinking about novel critical physiology in septic shock, hypoxemic respiratory failure, and cardiac dysfunction. But it allows a more standardized approach to analyzing data and telling a more complete picture.

Adrianne Rahde Bischoff (12:12.302)
And this is probably irrelevant for some of our audience, but in case we have younger learners who move fast, this is exactly the reason you cannot read just the abstract. If you go by the title and abstract only of a paper like this, you will conclude that norepinephrine and dopamine are equivalent for septic shock in neonates and it does not matter which you use. You have to read how the study was done, what the methods were, what the limitations are, and critically appraise the literature. Otherwise you will be led in the wrong direction and you will not make thoughtful decisions.

Nim Goldshtrom (12:50.100)
Because as you can tell from this one journal club alone, people have already made the mistakes and had them peer reviewed. You do not have to replicate them. You learn so much from getting into the nitty-gritty of what has already been done, and then you take it one step further and try to make it better with your next study. The beautiful part of research, as critical as we have tried to be for our audience in the context of hemodynamics, cardiac function, and congenital heart disease, is that these are solvable problems. It just takes time and investment and attention to detail. We hope this does not dissuade anyone from doing research. Our interpretation is not that people spend all this time and do not move the needle. They are moving the needle in ways that are not always obvious. It takes the mistakes to get to the successes. That is the fascinating and personally gratifying part of doing research. It is hard and it takes a long time. But to your point, Adrianne, you cannot read an abstract. You have to understand why the paper exists, and the methods section is probably the most important section, if not second most important, before everything else that comes from it.

Adrianne Rahde Bischoff (14:18.810)
Well, I think this has been a great journal club. We covered a lot of sepsis today, and I got lucky finding three articles all published within the last three months, which made for a very timely episode. I cannot wait to see what you come up with next.

Nim Goldshtrom (14:40.100)
I cannot wait either. I have a bunch of thoughts. I have a list of topics in the congenital heart space that are adjacent to hemodynamics, so it should be an interesting next journal club.

Adrianne Rahde Bischoff (14:50.100)
And let us know what you all think. If this is the format you enjoy, or if there is something else you would like us to discuss, we read all the comments and will try to tailor the show to what you want to hear from us.

Nim Goldshtrom (15:00.100)
This has been fun. Can't wait for the next time, Adrianne.

Adrianne Rahde Bischoff (15:03.100)
Yeah, take care, bye.

Nim Goldshtrom (15:08.582)
Bye everyone.