Invited Commentary 

Pediatrics

April 11, 2025

Vasoactive Selection for Pediatric Septic Shock—Where to Begin?

Scott L. Weiss

JAMA Netw Open. 2025;8(4):e254726. doi:10.1001/jamanetworkopen.2025.4726

Vasoactive medications are commonly used to reverse cardiovascular dysfunction and optimize tissue perfusion in children with fluid-refractory or fluid-averse septic shock. The 2020 Surviving Sepsis Campaign International Guidelines for the Management of Septic Shock and Sepsis-Associated Organ Dysfunction in Children suggest using epinephrine or norepinephrine, rather than dopamine, for children with septic shock.¹ The shift away from dopamine as the preferred first-line vasoactive medication in pediatric septic shock was spurred by 2 clinical trials demonstrating lower mortality and more organ failure-free days with initial treatment and titration of epinephrine compared with dopamine.^2,3 However, the outcomes of receiving epinephrine compared with norepinephrine have not yet been assessed in pediatric septic shock. Given the complex pathophysiology of cardiovascular dysfunction in sepsis that variably includes vasomotor instability, some clinicians may select norepinephrine instead of epinephrine as the first-line vasoactive medication.¹

In this issue of JAMA Network Open, Eisenberg et al⁴ observed first-line epinephrine to be associated with higher 30-day mortality than first-line norepinephrine (3.7% vs 0%; risk difference: 3.7%, 95% CI, 0.2% to 7.2%) in a propensity-matched analysis of 231 patients aged 1 month to 18 years who presented to a single, academic emergency department with septic shock. The primary outcome, major adverse kidney events within 30 days, was not significantly different between groups using a propensity score-based inverse probability treatment weighting analysis. Of note, patients with echocardiographic evidence of myocardial dysfunction were excluded, and the group treated with first-line epinephrine appeared to have a higher illness severity and began receiving vasoactive medications earlier (among other between-group differences). However, the propensity score model generally offered satisfactory balance among the measured covariates and there was reasonable overlap of propensity scores across groups.

These data provide the most robust comparison to date between first-line epinephrine and norepinephrine for pediatric septic shock. Although the effect size was small, the findings suggest a survival benefit for treatment with first-line norepinephrine among children with septic shock without known cardiac dysfunction. Notably, the findings align with a recent open-label randomized clinical trial of children with fluid-refractory septic shock and clinical signs of cold shock (cool extremities, delayed capillary refill time, diminished pulses, narrow pulse pressure) that found first-line norepinephrine (combined with dobutamine) led to a shorter time to shock resolution than first-line epinephrine.⁵ Mortality at 28 days, while not significant, was also lower in the group receiving norepinephrine and dobutamine (24% vs 39%; P = .16). Moreover, guidelines currently recommend norepinephrine as the first-line vasoactive agent for adult septic shock, albeit without clear evidence of improved mortality compared with epinephrine and with several caveats.⁶

While provocative and important, Eisenberg et al⁴ does not yet push the quality of available evidence to a clear, preferential first-line vasoactive strategy for children with septic shock. Despite appropriate statistical methods to account for between-group differences used by Eisenberg et al,⁴ it is difficult to remove the bias inherent in retrospective, observational studies. In this study,⁴ patients treated with first-line norepinephrine received a vasoactive medication later in their resuscitation course compared with those receiving epinephrine (median, 7.5 hours vs 3.6 hours after presentation) and much more often in the intensive care unit (65% vs 22%). Thus, it is possible that the norepinephrine group was enriched with patients who presented with lower severity of illness for whom vasoactive initiation could be delayed. This would allow more time to exclude septic myocardial dysfunction to identify a subpopulation for whom peripheral vasoconstriction with norepinephrine would be most appropriate. Conversely—or perhaps in addition—preferential early use of epinephrine, mostly initiated in the emergency department, may have conferred increased risk if patients were not adequately fluid resuscitated.⁷ Such factors are extremely difficult to disentangle from the exposure variable of first-line epinephrine or norepinephrine. The story of corticosteroid treatment in septic shock should provide a cautionary tale for which numerous observational studies demonstrate an association with mortality, while the totality of evidence from interventional trials favors benefit over harm.⁸

It is also reasonable to ponder the need to establish a preferred first-line vasoactive medication for septic shock. The pathophysiology of sepsis varies considerably between and within patients and the current state of cardiovascular dysfunction at any moment in time is an amalgamation of hypovolemic, cardiogenic, distributive, and dysoxic shock. Therefore, it is likely that the preferred treatment strategy also varies considerably between and within patients. Identifying an average treatment effect of epinephrine vs norepinephrine might increase the likelihood that a specific patient benefits from a single vasoactive medication, but a strategy that seeks to measure and serially reassess cardiovascular physiology no doubt offers the most personalized treatment option. Indeed, among patients evaluated with echocardiography at the same institution as Eisenberg et al,⁴ treatment with vasoactive strategies that were not appropriate for the measured cardiovascular physiology were associated with harm, while management changes that better supported the identified pathology led to improved outcomes.⁹ Ultimately, a one-size-fits-all approach, while more simple and pragmatic, may not produce the best outcome for everyone.

For now, and until further data are available, it remains reasonable to initiate either epinephrine or norepinephrine as a first-line vasoactive medication in children with septic shock without known cardiac dysfunction. Perhaps more important than the initial vasoactive agent is close attention to the physiologic response to that agent so that the chosen treatment plan is not blindly pushed forward without consideration of alternative approaches if the expected clinical improvement is not observed. After all, it is how you finish that counts, not how you start.