Acute respiratory distress syndrome (ARDS) accounts for almost 10% of intensive care units (ICU) admissions. Three ARDS stages have been defined, based on the PaO2/FIO2 ratio measured with positive end-expiratory pressure (PEEP) ≥5 cmH2O: mild (201-300 mmHg), moderate (200-101 mmHg), and severe (≤100mmHg). They represent 30.0%, 46.6%, and 23.4% of ARDS, respectively.

Mechanical invasive ventilation (MV), the cornerstone of ARDS patient care, has a primary goal to protect the lung from ventilator-induced lung injury (VILI). Delivering MV in a prone position (PP) has been shown to improve oxygenation, protect the lung through a better homogenization of lung stress/strain, and stabilize hemodynamics.

A meta-analysis of four randomized controlled trials showed beneficial effect of PP vs. supine position (SP) in the most hypoxemic patients. A fifth randomized controlled trial further showed a significant reduction in mortality with PP in ARDS patients with PaO2/FIO2 <150 mmHg, when neuromuscular blockade and long prone positioning sessions were used. Therefore, PP has since been strongly recommended for ARDS patients with PaO2/FIO2 <150 mmHg. Yet, there is limited evidence in patients with mild to moderate ARDS.

There are, however, strong arguments supporting the need for a new trial in ARDS patients with PaO2/FIO2 in the range 150-300 mmHg:

1. There is no trial that has specifically tested PP in this ARDS subset;
2. PP is safe and has become a standard of care in ICU;
3. Should VILI prevention be a mechanism through which PP improves survival, this should be involved in all ARDS patients;
4. The mortality at hospital discharge in this subset of ARDS remains significant, amounting to 34.9% (95% confidence intervals 31.4-38.5%) in mild and 40.3% (37.4-43.3) in moderate stages;
5. Among 580 patients with mild ARDS at admission to the ICU, in-hospital mortality was 10%, 30%, and 37% for those who improved, persisted, and worsened ARDS, respectively.
6. Finally, PP has been shown to be cost-effective under commonly accepted thresholds.

The hypothesis is that in patients within the 150-300mmHg PaO2/FIO2 range at the time of ARDS diagnosis, PP can reduce mortality as compared to a similar group left in the SP.

Inclusion Criteria:

• Age of 18 years or more admitted to ICU.
• ARDS patients meeting the Berlin criteria under invasive mechanical ventilation within the first 24h.
• Mechanically ventilated patients with standard settings: PEEP of 5 cmH2O or more and VT 6 ml/kg PBW and Pplat equal to or below 28-30 cmH2O after intubation.
• PaO2/FIO2 between 150 and 300 mmHg under previous settings within 12 hours after mechanical ventilation standardization.
• Written informed consent obtained from the patient, next of kin/proxy or emergency consent. The patient will be asked to give his consent for the continuation of the trial when his condition will allow.

Exclusion Criteria:

• Contra-indication to PP: spine instability; intracranial pressure greater than 20 mmHg; severe facial trauma; hemodynamic instability
• Non-inclusion criteria related to medical condition: Lung transplant; Burns over more than 20% of body surface; Chronic respiratory failure requiring home oxygen supplementation and/or non-invasive ventilation; Underlying disease with death expectancy within one year;
• Other non-inclusion criteria: Therapeutic limitation; inclusion into another interventional study in intubated and ventilated patients with mortality as primary end-point in the last 30 days; previously included in the same study; prone positioning before inclusion.
• Individuals referred to in Articles L. 1121-5 à L. 1121-7 (pregnant women, persons deprived of freedom, minors), adult individuals under protection of law (L. 1121-8 and L. 1122-1-2) (emergency situations) of public health regulation in France
• Patients not affiliated to social security insurance regimen