This study focuses on patients who are at risk of developing a serious, life-threatening respiratory disease called Acute Respiratory Distress Syndrome (ARDS), which severely disrupts the function of their lungs.

Preclinical studies have shown that the use of a volatile anesthetic agent such as Sevoflurane could be beneficial in the treatment and prevention of this respiratory condition. By improving gas exchange and attenuating pulmonary inflammation in particular, this agent would make it possible to prevent deterioration or to restore pulmonary function more rapidly.

Half of the patients will receive inhaled sedation with sevoflurane and the other half will receive intravenous sedation already routinely used in participating ICUs (typically propofol, dexmedetomidine or a benzodiazepine, i.e. drugs approved for sedation).

The aim of this study is to assess whether the use of Sevoflurane could be beneficial in the prevention of ARDS.

MAIN OBJECTIVE To assess the efficacy of inhaled sevoflurane, compared to current intravenous sedation practice, for improving PaO2/FiO2 in ICU patients at high risk for ARDS.

HYPOTHESIS The investigators hypothesized that a strategy of inhaled sedation with sevoflurane could be more effective than current intravenous sedation practice at improving pulmonary function during the early days of ICU admission, in patients at risk of ARDS.

Inclusion Criteria:

1. Age ≥ 18 years
2. Admitted to participating ICUs with at least one known risk factor for ARDS and a LIPS equals to, or greater than, 4 (Appendix D)105
3. Patient under invasive mechanical ventilation
4. With expected duration of sedation superior or equal to 4 hours
5. Affiliation to the French Sécurité Sociale

Exclusion Criteria:

• Patient under judicial protection, guardianship or supervision, as defined by art L1121-8 of the Public Health Code
• Patient under psychiatric care as defined by art. L1121-6 of the Public Health Code
• Patient deprived of their freedom by judiciary or administrative order
• Known pregnancy
• Presence of ARDS prior to randomization
• Endotracheal ventilation for greater than 24 hours prior to randomization
• Home mechanical ventilation (non-invasive ventilation or via tracheotomy) except for CPAP/BIPAP used solely for sleep-disordered breathing
• Tidal volume of 6 mL/kg predicted body weight (PBW) below 200 mL (i.e. height inferior to 134cm for a man and 139cm for a woman)
• Moribund patient, i.e. not expected to survive 24 hours despite intensive care
• Previous hypersensitivity or anaphylactic reaction to sevoflurane or to the intravenous sedation agent routinely used in the participating ICU (such as midazolam, propofol, or dexmedetomidine)
• Absolute contra-indications to the intravenous sedation agent routinely used in the participating ICU (such as midazolam, propofol, or dexmedetomidine)
• Medical history of malignant hyperthermia
• Long QT syndrome at risk of arrhythmic events
• Medical history of liver disease attributed to previous exposure to a halogenated agent (including sevoflurane)
• Suspected or proven intracranial hypertension
• Enrollment in another interventional trial with direct impact on oxygenation

Patients will be followed up for primary and secondary endpoints by members of the research staff who will be unaware of the trial group allocation. Information on whether the primary and secondary outcomes occur will be collected and entered into the electronic web-based case report form (eCRF) by trial or clinical trained personal (clinical research associate), blinded to the allocation group, under the supervision of the local principal investigator (PI) or designee who will also be unaware of the trial group allocation.

Finally, the independent trial statistician and the members of the data monitoring and safety committee (DMSC) will also remain blinded for the allocation during analysis. However, the observation of differences in serious adverse events between the two groups will allow, for safety reasons may the DMSC deem necessary, to unblind allocation groups.