The purpose of this study is to determine if a weaning strategy from artificial ventilation governs by respiratory behaviour status assessed by our method is safe enough.

Current guidelines for mechanical ventilation weaning do not apply for patients with altered level of consciousness. One major limitation is a failure of clinical assessment of the respiratory behaviour status of these patients that are not able to interact appropriately with the examiner. We propose a non-invasive method for the respiratory behaviour assessment of a patient under artificial ventilation that do not requires his participation. Our hypothesis is that, in case of normal respiratory behaviour status, it is possible to wean under security a patient despite his abnormal level of consciousness.We previously demonstrate that patients successfully wean by the clinical team also disclose a normal respiratory behaviour status while patients with weaning failure had abnormal respiratory behaviour. In the present study, we wonder to evaluate that a weaning strategy governs by respiratory behaviour status assessed by our method is safe. For this purpose, in this study, tracheal tube withdrawal is triggered by a recognition of a normal respiratory behaviour status assessed daily in patients under mechanical ventilation with an altered level of consciousness and a good tolerance to T tube challenge.

Inclusion Criteria:

• Altered level of consciousness (-1 < RASS > +1)
• Mechanical ventilation ( > 72 hours)
• Withdrawal of sedative drugs (> 48 hours)
• T tube challenge tolerance
• Signed approval for the study by close relative or legal representative

Exclusion Criteria:

• Respiratory disease prior to the actual artificial ventilation requirement.
• Impairment of upper airway function prior to the actual artificial ventilation requirement.
• Pregnancy
• Minor