Optimization of Positive End Expiratory Pressure by Use of Pulmonary Ultrasound for Patients With Blunt Chest Trauma Treated by Non-Invasive Ventilation (Opti-PEP).

This is an open-label, randomized controlled study comparing two non-invasive ventilation initiation strategies. Patients may be included if they are admitted to intensive care unit or continuing care unit and present with acute respiratory failure related to blunt chest trauma. We will not include patients with another indication for non-invasive ventilation, immediately needing invasive ventilation, a contraindication to non invasive ventilation or an estimated length of stay less than or equal to 48 hours.

Both groups will be treated with non-ivasive ventilation according to the medical prescription. The intervention group will benefit from the use by the physiotherapist of pulmonary ultrasound for the adjustment of PEEP during the 1st session. The conventional group will benefit from the non-invasive ventilation according to the usual care. The other non-invasive ventilation settings will be chosen with the same method in both groups.

The main endpoint will be the PaO2/FiO2 ratio (Arterial oxygen pressure / inspired fraction of oxygen) after 30 min of non-invasive ventilation treatment. This ratio is the best reflection of the patient oxygenation and currently used for respiratory therapies studies.

• Lung Ultrasound for PEEP setting (Other): NIV is started and adjusted by the physiotherapist. For the adjustment of PEEP, the ultrasound probe is placed on the thorax, facing the intercostal space having obtained the highest score during the initial LUS (synonym of the worst pulmonary aeration) and the images observed in real time. PEEP is increased until the physiotherapist cannot see any additional benefit on pulmonary aeration or appearance of a patient discomfort or an increasing of air leaks.
• Non invasive ventilation session (Procedure): The session will be done via a mouthpiece preferably, or a face mask if it is impossible for the patient to close their mouth properly or to hold the mouthpiece. The duration of the NIV session will be defined by the prescribing doctor (usually between 30 and 60 minutes), without being able to be less than 30 minutes. The session will be stopped in the event of the appearance of one of the contraindications to NIV mentioned above.
• Arterial blood gas (Biological): Arterial blood gas will be done at inclusion, at Visit 1 (at the 30th minute from the start of the 1st Non-invasive ventilation session) and at Visit 2 (at the 60th minute from the start of the 1st Non-invasive ventilation session) in both groups of patients in order to calculate the Pa02/Fi02 ratio.
• Interview on pain and comfort (Other): The investigator physiotherapist interviews the participant to measure pain and comfort scores using a visual analogue scale (score from 0 to 10).

Primary outcomes

• PaO2/FiO2 ratio (Arterial oxygen pressure / inspired oxygen fraction) in both groups [at the 30th minute from the start of the 1st Non-invasive ventilation session]: PaO2/FiO2 ratio (Arterial oxygen pressure / inspired oxygen fraction) in both groups. This PaO2/FiO2 ratio is now the benchmark for qualifying the severity of the respiratory impairment or measuring the response to a mechanical ventilation strategy aimed at improving oxygenation.

Secondary outcomes

• Comparaison of the Lung Ultrasound Score (LUS) in both groups [at the 30th minute from the start of the 1st Non-invasive ventilation session]: Comparaison of the Lung Ultrasound Score (LUS) in both groups. The LUS ranges from 0 to 36. Rib cage is divided into 12 areas (6 for each side) and each area is assessed according to the following scale: 0: Normal aeration corresponding to presence of lung sliding with A lines or fewer than two isolated B lines 1. Moderate loss of lung aeration corresponding to multiple well-defined B lines or spaced ultrasound lung called 'cornet-tail artifact" 2. Severe loss of lung aeration corresponding to multiple coalescent B lines or multiple abutting ultrasound lung cornet-tails issued from the pleural line 3. Lung consolidation corresponding to presence of a tissue pattern containing hyperechoic punctiform images representative of air bronchograms. Presence or absence of regional pulmanory blood flow and/or dynamic bronchograms.
• Comparaison of the PEEP value set at the start of the non-invasive ventilation session [At the start of the non-invasive ventilation session]: Comparaison of the PEEP value set (in cmH2O) in both groups
• Comparaison of the confort value in both groups [at the 30th minute from the start of the 1st Non-invasive ventilation session]: Comparaison of the confort value in both groups using the Confort Visual Analog scale (from 0 to 10)
• Comparaison of the pain value in both groups [at the 30th minute from the start of the 1st Non-invasive ventilation session]: Comparaison of the pain value in both groups using the Pain Visual Analog scale (from 0 to 10)
• Comparaison of the PaO2/FiO2 ratio in both groups [1 hour after the end of the 1st non-invasive ventilation session]: Comparaison of the PaO2/FiO2 ratio in both groups
• Comparaison of the incidence rate of respiratory aggravation in the 2 groups [Until discharge from intensive care/continuous care or until maximum D7.]: Comparaison of the incidence rate of respiratory aggravation in the 2 groups
• Rate of patients included in the experimental group in whom the use of ultrasound during the initiation of non-invasive ventilation is possible [at the 30th minute from the start of the 1st Non-invasive ventilation session]: Rate of patients included in the experimental group in whom the use of ultrasound during the initiation of NIV is possible
• PaO2/FiO2 ratio in both groups in patients with pulmonary consolidation [at the 30th minute from the start of the 1st Non-invasive ventilation session]: PaO2/FiO2 ratio in both groups in patients with pulmonary consolidation
• PaO2/FiO2 ratio in both groups in patients with pleural effusion [at the 30th minute from the start of the 1st Non-invasive ventilation session]: PaO2/FiO2 ratio in the 2 groups in patients with pleural effusion

Sex: All, ages: 18 Years to any

Inclusion Criteria:

* Patients aged 18 or more
* Admitted to intensive care or continuing care for 72 hours or less for blunt chest trauma, defined by the presence of at least one of the following elements on the initial CT scan: fracture(s) of rib(s) / fracture(s) of the sternum / pulmonary contusion / hemothorax
* Acute hypoxemic respiratory failure defined by the administration of oxygen at least 3 L/min with nasal cannula or FiO2≥30% if high flow oxygen therapy.
* Patient with a functional arterial catheter for blood tests

Exclusion Criteria:

* Acute respiratory distress, defined by the presence of at least one of the following clinical signs: respiratory rate ˃ 35 / use of accessory inspirators / paradoxal abdominal or thoracic motion
* Imminent need for invasive mechanical ventilation
* Usual contraindication to non-invasive ventilation (undrained pneumothorax, trauma to the face, vigilance disorders, digestive bleeding, hemodynamic instability, intolerance)
* Hypercapnia (PaCO2˃45mmHg)
* Patient unable to cooperate, communicate
* Therapeutic limitation
* Expected length of stay ≤ 48h
* Severe head trauma
* Pregnant or breastfeeding women
* Participation in other clinical research related to respiratory failure/respiratory therapy
* Vulnerable people
* Protected adults, under guardianship or curatorship, or unable to give consent
* Non-affiliated person or beneficiary of a social security scheme
* Absence of free, informed and written consent, signed by the participant and the investigator