Weaning from mechanical ventilation is a daily challenge in intensive care units, as it can take up to 50% of the total duration of ventilation. The longer the duration of ventilation is, the more there is complication related with it.

Even when the spontaneous breathing trial is succeeded, 10 to 20% of extubations are failed and requires re-intubation.

There is two different ways to assess if the patient is capable of breathing by its own : T-piece which can be considered as hard to succeed (it can delay extubation for some patients) or pressure support ventilation with no PEEP which can be too easy and lead to an extubation too early.

Studies have identified risk factors of weaning induced pulmonary oedema wich is one of the main cause of failed extubation (up to 60%).

The purpose of P-WEAN is to evaluate whether a personalized strategy for weaning from mechanical ventilation, including daily search for weaning criteria and individualization of the weaning modality (T-piece or pressure support ventilation with zero PEEP) based on the existence of WIPO risk factors (obesity, COPD, heart disease) improves weaning success compared with usual practice.

Before the study begin, all participating centers will be randomized to determine the time between the control and the interventional period. Patients included in the study will not be randomized individually but will be managed either according to the standard of care of the service or according to the personalized strategy (according to the stepped wedge group of the center).

• Control period: Patients will be subjected to a weaning strategy in accordance with the usual care practices of the service (evaluation of the criteria of weanability, the modalities of the spontaneous breathing trial and extubation)
• Interventional period: Patients will undergo a protocolized and standardized weaning strategy, including a daily assessment of weaning criteria, and a weaning trial (PSV or T-piece) whose modalities will be defined by the presence or the absence of weaning induced pulmonary oedema risk factors. These are :
• Obesity defined as BMI greater than 30 kg/m²
• Suspected or known COPD
• Heart disease whether it is structural (hypertrophic, dilated, valvular), functional (diastolic or systolic dysfunction), ischemic, or dysrhythmic.

Step 1 :

Daily assessment of weaning criteria in accordance with the study protocol

Step 2 :

• No risk factors : spontaneous breathing trial for 1 hour with pressure support ventilation (setting a pressure support level of 5 to 8 cm of water with no PEEP)
• Risk factors : spontaneous breathing trial with a T-piece for 1 hour, measuring protein and hemoglobin by blood sample before and after the trial.

Step 3 : Reconnecting the patient for at least 30 minutes and extubation if the SBT is successful. If not, the optimization of cardiac function or fluid overload will be the responsibility of the clinician.

Decisions regarding the use of non-invasive ventilation (NIV) or high-flow nasal oxygenation, will be made based on each centre's expertise and current clinical practice to minimize interference with the study procedure. However, knowing the recent data and the possible superiority of NIV in preventing post-extubation respiratory failure for high risk patients, investigators will be strongly encouraged to prioritize the prophylactic use of NIV (over high-flow nasal oxygenation) after tracheal extubation.

Patients will be assessed by members of the investigating center's research team (under the supervision of the centre's investigators) at least daily during hospitalization in the ICU and until day 28 post-randomization, as well as at discharge and at 90 days.

Inclusion Criteria : all patients ≥ 18 y.o admitted to the ICU, under invasive mechanical ventilation for more then 24 hours and fullfilling the following weaning criteria :

• Resolution or improvement of the condition that led to intubation, as judged by the clinician in charge of the patient.
• Hemodynamic stability, defined as systolic blood pressure between 90 and 160 mmHg, heart rate less than 140 beats/min, without or with low doses of vasopressors.
• Glasgow score of 13 or greater and Richmond Agitation-Sedation Scale (RASS) score between -1 and +1.
• Respiratory stability defined as: oxygen saturation > 90% with inspired oxygen fraction ≤ 0.5 and PEEP ≤ 8 cmH2O, respiratory rate < 35/min, spontaneous tidal volume > 5 mL/kg, and peak inspiratory pressure ≤ 15 cmH2O.
• Few secretions (< 3 aspirations in the past 8 hours).
• Effective cough.
• Negative leak test (>100 mL or >10%).
• No surgery planned within 72 hours.
• Patients with a social security plan.

Exclusion Criteria:

• Acute cerebral pathology requiring admission to intensive care and/or mechanical ventilation with tracheal intubation for neurological reasons (Glasgow score <13 at the time of intubation or CT scan abnormality): hemorrhagic stroke without or with vascular malformation (aneurysm, arterio-venous malformation...), ischemic stroke, head trauma, cerebral anoxo-ischemia after cardiac arrest.
• Tetraplegic or paraplegic patients with lesion level higher than D8.
• Peripheral neuromuscular pathology (underlying myopathy or myasthenia).
• ICU's Neuromyopathy.
• Tracheostomy.
• Patients with a decision of non-reintubation or terminal intubation.
• Pregnant or lactating women.
• Patients already included in this study.
• Patients under guardianship, curatorship or safeguard of justice.

Double blind is not possible as this study evaluate a care strategy. However, in order to limit evaluation bias, the methodologist will be blinded to the group.