Corticosteroids have moved from being one of the most controversial therapies in ARDS to an accepted treatment in selected patients. The 2026 review by Daoud, Villar and Annane argues that the key question is no longer simply whether corticosteroids work, but which patients are most likely to benefit and when treatment should be started.
Importantly, this publication is a narrative review, not a new guideline. It brings together randomised trials, meta-analyses and current guideline recommendations.
Why does timing matter?
ARDS evolves from an early inflammatory phase into proliferative and, in some patients, fibrotic lung injury.
Corticosteroids may be most useful while inflammation remains modifiable. This helps explain why earlier treatment has generally produced more favourable results than very late initiation.
The review highlights a key safety signal from persistent ARDS studies: corticosteroids started more than 14 days after ARDS onset were associated with possible harm. Routine initiation this late is therefore not supported.
What do current guidelines say?
The review cites the 2024 SCCM focused update and the ATS ARDS guideline, both of which support corticosteroids in ARDS with a conditional recommendation.
That conditional wording reflects uncertainty over the choice of corticosteroid, optimal dose, timing, duration and which patients benefit most.
This should not be interpreted as a recommendation to treat every patient with ARDS.
Which regimens are best supported?
The review highlights the following regimens by clinical setting. It also stresses that hydrocortisone, dexamethasone and methylprednisolone should not automatically be considered interchangeable. There are no head-to-head randomised trials proving that one is superior or equivalent across all ARDS settings.
| Clinical setting | Regimen highlighted in the review |
|---|---|
| Moderate-to-severe non-COVID ARDS | Dexamethasone 20 mg IV daily for 5 days, then 10 mg daily for 5 days; or methylprednisolone about 1–2 mg/kg/day with taper |
| COVID-19 requiring respiratory support | Dexamethasone 6 mg daily for up to 10 days |
| Severe community-acquired pneumonia | Hydrocortisone 200 mg/day |
| Septic shock with ARDS | Hydrocortisone 200 mg/day IV |
| Influenza-associated ARDS | Routine corticosteroids not recommended |
Severe community-acquired pneumonia
The CAPE COD trial provided strong evidence for hydrocortisone 200 mg/day in severe community-acquired pneumonia, reducing mortality and the need for intubation and vasopressors.
However, CAPE COD was not an ARDS-specific trial, and not all participants fulfilled ARDS criteria. Its findings therefore support hydrocortisone in severe CAP rather than proving benefit in every case of CAP-associated ARDS.
Septic shock
In patients with persistent vasopressor-dependent septic shock, hydrocortisone 200 mg/day IV has an established evidence base independent of the ARDS diagnosis.
Septic ARDS may also overlap with a more hyperinflammatory biological profile, although this is not yet a validated treatment-selection strategy.
COVID-19
COVID-19 transformed the corticosteroid evidence base. Dexamethasone 6 mg daily for up to 10 days is strongly supported in patients requiring oxygen or ventilatory support.
This benefit should not automatically be extrapolated to other viral causes of ARDS.
Why is influenza different?
Influenza remains an important exception.
Observational studies have associated corticosteroids with higher mortality, secondary infection and longer ICU stay, although confounding by indication is a major limitation.
Because robust randomised evidence of benefit is still lacking, the review advises against routine corticosteroid use in influenza-associated ARDS unless another accepted indication exists, such as septic shock, asthma or COPD exacerbation, or adrenal insufficiency.
What about persistent ARDS?
Corticosteroids may still be considered in selected patients with persistent inflammatory or organising lung injury, but the evidence is much weaker than in early ARDS.
The key caution is timing: routine initiation after day 14 is not supported.
Imaging findings and biomarkers such as PIIINP have been proposed to identify ongoing fibroproliferation, but none has been prospectively validated for routine corticosteroid selection.
Could inflammatory phenotypes guide treatment?
ARDS is increasingly recognised as biologically heterogeneous.
A hyperinflammatory phenotype, characterised by higher inflammatory markers, greater shock and higher mortality, may derive more benefit from corticosteroids than a hypoinflammatory phenotype.
However, this evidence comes mainly from secondary and retrospective analyses. Phenotype-guided corticosteroid treatment is therefore not ready for routine clinical use.
What adverse effects matter?
The most consistent adverse effects are hyperglycaemia and hypernatraemia.
Secondary infection, ICU-acquired weakness and myopathy remain important clinical concerns, particularly with prolonged treatment or additional immunosuppression.
Monitoring should include glucose, electrolytes, infection risk and neuromuscular recovery.
Clinical takeaway
Corticosteroids are an important treatment in selected patients with ARDS, but they should not be used as a one-size-fits-all therapy.
The strongest current evidence supports treatment in severe COVID-19, severe community-acquired pneumonia and septic shock, while broader non-COVID ARDS carries a conditional recommendation.
For everyday practice, the two most important considerations are aetiology and timing. Early treatment appears more favourable, while routine initiation beyond 14 days should be avoided.
Biomarker- and phenotype-guided treatment may eventually allow more precise corticosteroid use, but these approaches remain investigational.
Full reference
Daoud T, Villar J, Annane D. Corticosteroids in ARDS: old controversies, new insights, and future directions. Intensive Care Medicine. Published 13 August 2026. doi:10.1007/s00134-026-08567-3.