The care of acute respiratory distress syndrome (ARDS) has been significantly improved by learning from experimental and physiological research works and by a series of randomized controlled trials. The mortality of this syndrome remains however high. Numerous experimental and clinical works demonstrated that a ventilatory mode authorizing the patient to make, from the acute phase, spontaneous breathing cycles superimposed on assistance delivered by the ventilator (BIPAP-APRV mode) allowed to improve gas exchanges and hemodynamic tolerance of the ventilation while reducing the need for sedative drugs. This ventilatory mode could also reduce the risk of diaphragmatic dysfunction induced by ventilation. Consequently, our hypothesis is that this ventilatory mode could allow a reduction of mortality in ARDS patients.

The aim of this multicenter, prospective, randomized, controlled, open study is to compare the effects of two ventilatory strategies on the mortality of ARDS patients and placed under mechanical ventilation.

Inclusion Criteria:

• Within 1 week of a known clinical insult or new or worsening respiratory symptoms
• Intubation and mechanical ventilation
• Bilateral opacities - not fully explained by effusions, lobar/lung collapse, or nodules lung
• Report PaO2 / FiO2 ≤ 200 mmHg with a PEP of at least 5 cmH2O
• Respiratory failure not fully explained by cardiac failure or fluid overload
• Criteria 1, 2 and 3 presents jointly for less than 48 hours
• Consent to participate obtained either from the patient himself or from a relative.

Exclusion Criteria:

• Age < 18 years
• Pregnancy
• Predictable duration of the mechanical ventilation on endotracheal tube lower than 48 hours
• Patient being in period of exclusion further to the participation in another biomedical study
• intracranial hypertension (suspected or confirmed)
• known or suspected COPD
• Chronic respiratory failure treated by long-term oxygen and / or long-term respiratory support
• Morbid obesity defined as weight greater than 1 kg / cm
• Sickle Cell Disease
• Marrow recent transplant, post-chemotherapy aplasia
• Widened burns (> 30% body surface area)
• Severe hepatic cirrhosis (Child-Pugh C)
• Pneumothorax (drained or not)
• Treatment with extracorporeal support (ECMO)
• Decision of active therapeutic limitation
• Unavailability of the model of respirator that must be used in the study
• Failure to obtain a consent by persons authorized to do so.
• Patient under law protection.
• Person non-beneficiary of a social security system