During preoxygenation, imperfect seal between the face mask and patient's face can induce an inward air leak decreasing its effectiveness. We assume that noninvasive ventilation could cancel the effect of the leak.

This is a prospective study. Healthy volunteers are randomised in cross-over between spontaneous breathing or noninvasive ventilation pressure support preoxygenation in the presence or absence of a calibrated leak on the inspiratory circuit.

During preoxygenation, imperfect seal between the face mask and patient's face can induce an inward air leak decreasing its effectiveness. We assume that noninvasive ventilation could cancel the effect of the leak.

We planned an experimental study with healthy volunteers (residents in anesthesiology).

Healthy volunteers are randomised in cross-over between preoxygenation through spontaeous breathing or noninvasive ventilation pressure support preoxygenation in the presence or absence of a calibrated leak on the inspiratory circuit. The inspiratory fraction of oxygen is 100%. The exhaled gas (oxygen and carbon dioxide) are monitored.The subjects breathed through a mouthpiece (with a nose clip) connected to an anesthetic ventilator.

The primary endpoint is the time to achieve end expiratory fraction of oxygene of 90% or more.

The secondary endpoint is the proportion of subject with end expiratory fraction of oxygen of 90% or more after a 3 min period (recommended in clinical guidelines).

Inclusion Criteria:

• healthy volunteers

Exclusion Criteria:

• any pathological condition
• active tobaco