Burns affect more than 11 million people worldwide each year. These injuries are responsible for severe morbidity resulting in a high societal burden and account for more than 180,000 yearly deaths especially in low- and middle-income countries.

Major burns induce an important local and systemic inflammatory response that may be overwhelmed. This inflammation is a physiological phenomenon that favours the healing of tissues. However, the overproduction of inflammatory mediators might lead to an exacerbated Systemic Inflammatory Response Syndrome (SIRS). Recently the total body surface area (TBSA) burned has shown to be well correlated to persistent elevation of pro-inflammatory mediators (such as IL-6). This SIRS, in turn, contributes to the enhanced risk of sepsis, acute respiratory distress syndromes (ARDS) and organ failures in general such as acute kidney injuries (AKI), most of those occurring within the first week of admission.

Corticosteroids (CS) have already proven their effectiveness against SIRS-induced organ dysfunction or mortality in acute medicine notably in septic shock, polytraumatized patients and more recently in the treatment of viral or non-viral ARDS without increasing the risk of secondary bacterial complications or significant side effects . Indeed the recent SCCM Guidelines clearly advocate for the use of CS in severe community-acquired pneumonia, septic shock and ARDS. The investigators recently performed a large multicenter, double-blinded randomized controlled trial (the PACMAN trial, PHRC-N 2016) including 1222 patients scheduled for major surgery in which the investigators observed a major decrease in CRP blood concentrations in the dexamethasone arm. The rate of AKI and the need for mechanical ventilation were also significantly reduced in the intervention arm. ICU Patients with severe burns undergo several surgeries, including major procedures (excision, skin grafts), rendering them quite similar to those in the PACMAN trial in terms of inflammatory response. Very few side effects (hyperglycemia mainly) easily overcome in ICU are usually reported with the use of low-to-moderate dose of CS.

In severe burn patients, very few data are available to date, two retrospective case control studies and a small prospective randomized trial showed promising results when using CS but high quality evidence is lacking.

The investigators hypothesise here that the use of dexamethasone after major burns, the prototypic model of inflammatory response in surgical ICU patients, would limit SIRS-induced organ failure and/or all-cause mortality.

Inclusion Criteria:

• 18 years old ≤ Age ≤ 80 years old.
• Total burn surface area ≥ 20%, measured by a trained expert upon admission
• Invasive mechanical ventilation at the time of inclusion
• Within 48 hours of the burn injury
• Informed and signed written consent of the next-of-kin, legal representative (trusteeship, guardianship) or emergency procedure in the absence of a legal representative.
• Affiliation with French social security system or beneficiary from such system

Exclusion Criteria:

• Imminent death and a do-not-resuscitate order

  • Medical history of hypersensitivity to dexamethasone and hypersensitivity to all of its excipients
  • Pregnancy (attested by a pregnancy test for women of childbearing age) and/or breastfeeding women
  • Participation to another interventional study involving a drug with known interactions with dexamethasone
  • Uncontrolled viral hepatitis or invasive fungal infection at the time of inclusion
  • Prolonged administration of steroids in the last 90 days (>0.3 mg/kg/day of equivalent prednisolone)
  • Moderate-to-severe ARDS upon admission (according to Berlin definition criteria)