New England Journal of Medicine February 25, 2021

Dexamethasone in Hospitalized Patients with Covid-19

RECOVERY Collaborative Group

Bottom Line

The RECOVERY trial demonstrated that low-dose dexamethasone reduces 28-day mortality in hospitalized patients with COVID-19 who require respiratory support, but provides no benefit and possible harm in those who do not require oxygen.

Key Findings

1. Among all randomized patients, dexamethasone resulted in a significantly lower 28-day mortality compared to usual care (22.9% vs. 25.7%; rate ratio, 0.83; 95% CI, 0.75 to 0.93; P<0.001) [4.1.8].
2. The survival benefit was most pronounced in critically ill patients requiring invasive mechanical ventilation at randomization, with dexamethasone reducing mortality from 41.4% to 29.3% (rate ratio, 0.64; 95% CI, 0.51 to 0.81).
3. A smaller but statistically significant mortality reduction was observed in patients receiving oxygen without invasive mechanical ventilation (23.3% vs. 26.2%; rate ratio, 0.82; 95% CI, 0.72 to 0.94).
4. Dexamethasone provided no survival benefit, and suggested potential harm, in patients who were not receiving any respiratory support at the time of randomization (17.8% vs. 14.0%; rate ratio, 1.19; 95% CI, 0.92 to 1.55).

Study Design

Design
RCT
Open-Label
Sample
6,425
Patients
Duration
28 days
Median
Setting
Multicenter, UK
Population Hospitalized patients with clinically suspected or laboratory-confirmed SARS-CoV-2 infection
Intervention Oral or intravenous dexamethasone 6 mg once daily for up to 10 days
Comparator Usual standard of care
Outcome 28-day all-cause mortality

Study Limitations

The open-label nature of the trial could theoretically introduce bias in the delivery of ancillary care or the decision to withdraw life-sustaining treatments, although all-cause mortality is an objective and robust endpoint.
The rapid, pragmatic design meant that detailed baseline laboratory variables, inflammatory markers, and long-term functional recovery data beyond 28 days were not systematically collected.
The timing of symptom onset was self-reported, which could confound the exact demarcation between the early viral replication phase and the later hyperinflammatory phase.
The exceptionally high mortality rates observed in the control group reflect the severity of the first pandemic wave in the UK, which may differ from current baselines influenced by widespread immunity and contemporary viral variants.

Clinical Significance

The RECOVERY trial fundamentally changed the standard of care for severe COVID-19 globally by proving that systemic corticosteroids reduce mortality. Crucially, the trial established a stratified treatment paradigm: immunomodulation is highly beneficial for patients in the hyperinflammatory phase (requiring oxygen or ventilation) but is contraindicated during the early viral phase for patients not requiring supplemental oxygen.

Historical Context

During the first wave of the COVID-19 pandemic in early 2020, mortality among hospitalized patients—especially those mechanically ventilated—approached devastatingly high levels. There were no proven targeted therapies, and the use of corticosteroids for viral pneumonia was highly controversial due to historical data from SARS, MERS, and influenza suggesting potential harm and delayed viral clearance. RECOVERY, a rapidly deployed adaptive platform trial in the UK, provided definitive evidence within months that overturned previous hesitation and identified the first life-saving therapeutic intervention for the disease.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

Why does dexamethasone provide a mortality benefit for COVID-19 patients requiring oxygen but potentially cause harm in those breathing ambient air?

Key Response

This highlights the biphasic nature of severe COVID-19. Early on, the disease is driven by viral replication, and immunosuppression via dexamethasone can impair viral clearance. Later, in patients requiring oxygen, the disease is primarily driven by a hyperinflammatory immune response (cytokine storm), which dexamethasone effectively dampens, improving outcomes.

Resident
Resident

A patient admitted with COVID-19 has an SpO2 of 93% on room air but feels highly dyspneic. Based on the RECOVERY trial, should you initiate dexamethasone, and how does this decision impact your overall management strategy?

Key Response

The RECOVERY trial specifically showed benefit only for patients receiving respiratory support (oxygen or ventilation). Starting dexamethasone on a room-air patient goes against the evidence and risks harm. The resident must decide whether to start oxygen (which then qualifies them for dexamethasone) or hold off, emphasizing strict adherence to evidence-based clinical criteria rather than subjective dyspnea alone.

Fellow
Fellow

How does the timing of dexamethasone initiation relative to symptom onset influence its efficacy, and how do we reconcile the RECOVERY data with the use of other immunomodulators like tocilizumab or baricitinib in the ICU?

Key Response

The trial demonstrated greatest benefit in those with symptom duration over 7 days, aligning with the onset of the inflammatory phase. Fellows must integrate this with data on IL-6 inhibitors and JAK inhibitors, understanding that dexamethasone is the foundational therapy upon which further targeted immunosuppression is layered based on inflammatory markers and clinical trajectory.

Attending
Attending

Given the open-label nature of the RECOVERY trial and the massive paradigm shift it caused, how do we balance the rapid adoption of highly pragmatic, unblinded platform trial results with the historical gold standard of double-blinded RCTs in our clinical teaching?

Key Response

Attendings must navigate teaching evidence-based medicine in a pandemic era. The RECOVERY trial was pragmatic and open-label, which expedited enrollment and real-world applicability but introduced potential care biases. The insight here is recognizing when the magnitude of benefit (mortality) and the urgency of the situation validate adopting pragmatic trial designs over delayed, traditional double-blind trials.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The RECOVERY trial utilized a pragmatic, adaptive platform design with an unblinded structure. How does this design impact the statistical power and the potential for co-intervention bias, and what methods can be used to control for type I error across multiple treatment arms?

Key Response

The platform design allows multiple therapies to be tested simultaneously against a shared control group, increasing statistical efficiency. However, the unblinded nature risks co-intervention bias where supportive care differs based on group assignment. Researchers must discuss methods like alpha spending functions or hierarchical modeling used to manage multiple comparisons in adaptive trials.

Journal Editor
Journal Editor

As a peer reviewer, what primary threats to internal validity would you flag regarding the pragmatic enrollment process and the lack of standardization in baseline standard of care across the massive network of RECOVERY trial sites?

Key Response

An editor would critically evaluate the heterogeneity of care. With thousands of patients across numerous NHS hospitals, standard of care varied wildly. While this increases external validity (generalizability), the lack of granular data on concurrent treatments (like remdesivir or anticoagulation protocols) and baseline comorbidities could mask confounding variables and threaten internal validity.

Guideline Committee
Guideline Committee

Based on the RECOVERY trial, how should living clinical practice guidelines grade the recommendation for dexamethasone in hospitalized COVID-19 patients, and what specific guardrails must be placed in the guideline text regarding patients not requiring oxygen?

Key Response

The committee would assign a strong recommendation with high-quality evidence (Level A) for using 6mg of dexamethasone daily for up to 10 days in patients on supplemental oxygen or mechanical ventilation. Crucially, the guidelines (such as NIH or WHO COVID-19 guidelines) must explicitly issue a strong recommendation AGAINST its use in patients not requiring oxygen due to the signal of harm, emphasizing stewardship of immunosuppressants.

Clinical Landscape

Noteworthy Related Trials

2020

SOLIDARITY Trial

n = 11,330 · NEJM

Tested

Remdesivir, Hydroxychloroquine, Lopinavir, or Interferon

Population

Hospitalized adults with COVID-19

Comparator

Standard of care

Endpoint

In-hospital mortality

Key result: None of the evaluated repurposed drugs reduced mortality, initiation of ventilation, or hospital duration.
2020

ACTT-1 Trial

n = 1,062 · NEJM

Tested

Remdesivir

Population

Hospitalized adults with lower respiratory tract involvement from COVID-19

Comparator

Placebo

Endpoint

Time to recovery

Key result: Remdesivir was superior to placebo in shortening the time to recovery in hospitalized adults with COVID-19.
2021

REMAP-CAP Trial

n = 803 · NEJM

Tested

Tocilizumab or Sarilumab

Population

Critically ill patients with COVID-19 receiving organ support

Comparator

Standard of care

Endpoint

Respiratory and cardiovascular organ support-free days

Key result: Interleukin-6 receptor antagonists improved outcomes and reduced mortality in critically ill patients.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis