- Design
- multicentre prospective observational cohort study, 2021-2023
- Population
- 5,119 children under 18 admitted with suspected community-acquired sepsis via 11 emergency departments
- Primary outcome
- corticosteroid agent, dose and factors associated with administration in the first 24 hours
- Effect
- 9.6% overall and 31.3% of those meeting Phoenix criteria; dexamethasone 69.5%, hydrocortisone 25.4%
The Sepsis Epidemiology in Australian and New Zealand Emergency Departments study followed 5,119 children admitted through 11 emergency departments with suspected community-acquired sepsis between 2021 and 2023. Median age was 1.6 years. In-hospital mortality was 1.0% overall, and 12.7% among the 252 children who met Phoenix sepsis criteria.
Corticosteroids were given within the first 24 hours to 9.6% of all these children and to 31.3% of those meeting Phoenix criteria. Dexamethasone was the agent in 69.5% of cases and hydrocortisone in 25.4%, with marked variation between sites. Use was associated with older age, more than one comorbidity, a raised initial lactate, large-volume fluid resuscitation and any vasopressor.
The authors' central observation is that the great majority of this use fell outside fluid-refractory or catecholamine-refractory shock - the one indication guidelines support, and the one in which trial evidence exists. Some of this is legitimate: dexamethasone has its own indications in croup, meningitis and asthma, and a child with suspected sepsis may be receiving it for one of those. But a two-to-one preference for dexamethasone over hydrocortisone, and site-to-site variation of this size, is not explained by indication alone. This is descriptive data with no outcome comparison, so it does not show harm. What it shows is a practice pattern operating ahead of its evidence, and that is worth auditing locally before the next child receives a dose out of habit.
- Reserve corticosteroids in paediatric sepsis for fluid- or catecholamine-refractory shock.
- Record the indication when giving dexamethasone to a child with suspected sepsis - croup, meningitis and asthma are different decisions.
- Hydrocortisone, not dexamethasone, is the agent studied in refractory septic shock.
- Audit your own department's use; between-site variation here was large.
- This is descriptive practice data and does not show that steroid use caused harm.
Why it matters
A drug is being given to nearly a third of the sickest septic children largely outside the only indication that has been tested.
Don't overread it
Descriptive cohort data with no outcome comparison - it documents a practice pattern, not a harm.
The statistics, in plain English
This is a descriptive cohort with no comparison of outcomes between treated and untreated children, so nothing here says steroids helped or harmed. The factors associated with use - lactate, fluid volume, vasopressors - describe sicker children, which is exactly the confounding by indication that would make any outcome comparison in this dataset uninterpretable. The variation between sites is the most informative number, because sites do not differ in their children.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for paediatrics, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free