Insufficient recruitment of eligible patients.

Unresolved ARDS is defined by the persistence of ARDS criteria at the end of the first week of evolution despite an appropriate treatment of the cause of ARDS. A persistent ARDS is associated with an increased mortality and prolonged lengths of mechanical ventilation, ICU stay and hospitalization. Persistent ARDS is characterized by ongoing inflammation, parenchymal-cell proliferation, and fibroproliferation leading to disordered deposition of collagen. All of these pathways may be responsive to corticosteroid therapy.

Only two randomized controlled double-blinded trials assessed the use of corticosteroids for persistent ARDS. In 24 patients, Meduri et al. reported an improvement of lung function and survival (1). In 180 patients, Steinberg et al showed no effect of corticosteroids on survival (2). A lower risk of death was observed when corticosteroids were started before 14 days after the onset of ARDS (2).

Alveolar procollagen III is validated as a biomarker of active fibroproliferation. Alveolar procollagen III > 9 µg/L is associated to fibroproliferation (3).

As mortality was lower in patients who received corticosteroids while presenting a high alveolar level of procollagen III on inclusion, Steinberg et al. showed that patients presenting with a low level of procollagen III and treated with corticosteroids had an increased risk of death (2).

Investigatores hypothesize that the use of procollagen III could improve personalized decision-making regarding steroid treatment in patients presenting with persistent ARDS. The future of anti-fibrotic treatment, including corticosteroids, in persistent ARDS might propose to individualize the therapy according to the presence of an active fibroproliferative phase (precision or personalized medicine).

Inclusion Criteria:

• Age >= 18 years
• Continuous endotracheal ventilation
• Moderate - severe ARDS according to Berlin definition with PaO2/FiO2 ≤ 200 with PEEP >= 5 cm H2O
• Date of ARDS onset : >= day 5 and ≤ day 14 after the onset of ARDS criteria (regardless of ARDS severity)
• Procollagen III above 9 µg/L in a bronchoalveolar lavage performed by the attending physician between day 3 and day 13 after the onset of ARDS and realized within 5 days prior to randomization

Exclusion Criteria:

• Known pregnancy or breast feeding
• Participation to another interventional trial within 30 days with mortality or ventilator free days as the main endpoint
• Clinical evidence of active untreated infection
• A known, undrained abscess
• Intravascular nidus of infection
• Disseminated fungal infection