Skip to content

Preprint Data from RECOVERY Trial Show Mortality Benefit of Dexamethasone in COVID-19

The FLARE Four

  • As part of the RECOVERY trial, 2,104 patients with COVID-19 received dexamethasone 6 mg/day for up to 10 days and 4,321 received usual care
  • According to a preprint manuscript, among patients receiving respiratory support, steroid use was associated with a 3% absolute reduction in the risk of 28-day mortality, with greater benefit for patients on mechanical ventilation than those on oxygen
  • There was a nonsignificant trend toward harm in patients not requiring oxygen therapy who received steroids
  • Physicians should strongly consider the use of dexamethasone for COVID-19 patients who are on oxygen or have hypoxemic respiratory failure

According to recently released preprint data (not yet peer-reviewed) from the RECOVERY trial, dexamethasone has a mortality benefit in COVID-19 patients who require oxygen or mechanical ventilation. In a review posted on June 27, 2020, the Fast Literature Assessment and Review team at Massachusetts General Hospital puts the trial in context.


The role of steroids in acute respiratory distress syndrome (ARDS) and other critical illness syndromes is highly controversial. There is evidence suggesting the innate immune response is itself a mechanism of ARDS—it potentially gives way to maladaptive hyperinflammation—so there has been longstanding interest in immunomodulatory therapies.

Widespread adoption of such strategies has been limited by contradictory data. Two large randomized, clinical trials of different steroid regimens applied at different times in the ARDS course showed conflicting efficacy results. Other studies showed a signal for harm in late ARDS (>14 days) and ARDS due to viral pneumonia. Among survivors of ARDS, steroids have been associated with delirium and profound, long-lasting neuromuscular weakness.


RECOVERY is a multi-arm, multi-platform clinical trial testing a number of therapies for COVID-19 in parallel with 28-day mortality as a primary endpoint. The sample is massive: in the steroid arm, 2,104 patients received dexamethasone 6 mg/day for up to 10 days. They were compared with 4,321 patients who received usual care.


RECOVERY showed a 3% absolute mortality risk reduction in the steroid group among patients receiving respiratory support, with greater benefit in patients on mechanical ventilation than in those on oxygen. There was a nonsignificant trend toward harm in patients not requiring oxygen therapy who received steroids.


These data from RECOVERY should be practice-changing. However, given concerns about both short- and long-term adverse events with steroid use, more precise phenotyping is necessary. Clinicians must remain alert to the possibility that the groups that appeared to benefit in RECOVERY (those on oxygen or mechanical ventilation) are not the true subtypes that distinguish steroid responders from nonresponders.

In addition, any open-label trial is potentially confounded. Despite the new standard of care of providing dexamethasone to patients with COVID-19, there is a continued need for a fully blinded randomized, controlled trial.

View all COVID-19 updates

Learn more about pulmonary research at Mass General

Related topics


Within the next year, physicians can expect to see increased numbers of patients with fibrotic lung disease that has developed as a complication of acute respiratory distress syndrome related to COVID-19.


Massachusetts General Hospital has used veno-venous extracorporeal membrane oxygenation to support several carefully selected COVID-19 patients to recovery.