The aim of the study is to compare two devices for oxygen therapy, nasal high-flow and Venturi mask, in critically ill patients in the post-extubation period. The hypothesis is that nasal high-flow may be superior to the Venturi mask in terms of oxygenation

Acute respiratory failure is one the most common reason for admission and mechanical ventilation is the most frequent procedure performed in the intensive care unit (ICU). After extubation, oxygen therapy is commonly used to correct residual impairment in oxygenation. Oxygen therapy is usually performed via a Venturi mask allowing to deliver predetermined oxygen concentrations. The face mask is used in the place of nasal cannula in part because patients with acute respiratory failure (ARF) breath preferentially through an open mouth rather than the nose. The mask may, however, reduce comfort and patients with ARF, who are often agitated and poorly cooperative, may displace or self-remove the mask. Given that oxygen delivered to the patient is dry, clinical practice guidelines recommend humidifying the oxygen when above 4 l/min in the ICU setting, because the humidification function of the nasal mucosa can be insufficient at high oxygen flow rates and/or the critically ill patient with ARF often breathes through the mouth. Breathing dry oxygen could provoke dryness of the mouth, nose, throat and respiratory tract, resulting in discomfort and pain that are frequent in the ICU setting. Breathing dry air by the nose may also lead to the alteration of the mucociliary transport system and cause an increase of airway resistance in healthy subjects. However, there are no recommendations concerning the type of humidification device to use.

High-flow oxygen therapy is commonly practiced in the ICU. Recently, a method was described in which high-flow oxygen is delivered through a nasal cannula with an active humidification system that optimizes oxygen administration in patients with ARF (Optiflow, Fisher & Paykel, New Zealand). This system has several theoretical advantages. The delivery of high flows decreases dilution of the inhaled oxygen and, by matching patient's peak flow, allows accurate delivery of the set FiO2 throughout the whole inspiratory phase. In addition, a flow-dependent effect of continuous positive airway pressure, possibly due to an air entrainment mechanism, has been documented in healthy volunteers and in patients with COPD. Lastly, the use of high humidity levels may prevent damage to the ciliated epithelium of the airways, maintain mucociliary system activity, and facilitate the elimination of secretions. Through these effects, this new device for high-flow oxygen therapy has the potential to be effective and to improve tolerance to the treatment more than conventional systems for oxygen therapy, such as the Venturi mask.

In the present randomized, controlled trial, the investigators will compare two devices for oxygen therapy, nasal high-flow oxygen therapy and the standard Venturi mask, in critically ill patients who need oxygen in the post-extubation period. The hypothesis is that nasal high-flow is superior to the Venturi mask in terms of oxygenation

Inclusion Criteria:

• Mechanical ventilation > 24h
• Successful spontaneous breathing trial conducted for a period of 30-120 min.
• PaO2/FiO2 ≤ 300 at the end of the spontaneous breathing trial preceding extubation

Exclusion Criteria:

• age<18 years
• pregnancy
• tracheostomy
• need of NIV post-extubation (prophylactic NIV)