Evaluate the performance for measuring inspiratory effort of non-invasive mandibular movement analysis compared to the reference technique oesophageal pressure (PES) variation, in ventilated and spontaneously breathing Intensive Care Unity (ICU) patients, during weaning from mechanical ventilation and within 48 hours after extubation.

The investigators hypothesis is that the assessment of respiratory effort by MM analysis could represent a non-invasive and reliable alternative to the measurement of PES in critically ill patients.

Acute respiratory failure is the leading cause of ICU admission and is associated with high mortality. In patients with spontaneous breathing, inspiratory effort reflects ventilatory drive, which is frequently altered in critical illness due to the underlying pathology or its management.

Excessive inspiratory effort is associated with failure of non-invasive oxygenation strategies, difficult ventilator weaning, and increased mortality. Clinicians must therefore balance the benefits of maintaining spontaneous ventilation with the principles of protective ventilation to prevent lung injury induced either by the ventilator or by the patient.

The reference method for assessing inspiratory effort is esophageal pressure measurement (ΔPES), an invasive technique that limits routine clinical use. Recent studies suggest that non-invasive monitoring of mandibular movements provides a reliable assessment of respiratory effort.

The investigators hypothesize that analysis of mandibular movements could represent a reliable and non-invasive alternative to esophageal pressure measurement for evaluating respiratory effort in critically ill patients.

Inclusion Criteria:

• Adult patients (≥ 18 years)

• Admitted to the Intensive Care Unit for acute respiratory failure

• On invasive mechanical ventilation for more than 24 hours with:

  • A respiratory rate ≤ 35 breaths/minute;
  • Adequate oxygenation, defined as either an oxygen saturation of at least 90% (obtained during ventilation with a fraction of inspired oxygen [FiO2] ≤ 40% and a positive end-expiratory pressure [PEEP] ≤ 8 cm of water) or an arterial partial pressure of oxygen (PaO2) (measured in mmHg) / FiO2 ratio > 150 (obtained during ventilation with a PEEP ≤ 8 cm of water);
  • Respiratory muscle capacity allowing for an effective cough;
  • A state of wakefulness, defined as a Richmond Agitation and Sedation Scale (RASS) score of -2 to +1;
  • No use of continuous sedation;
  • No use of vasopressors (or use of minimal doses).

• At high risk of weaning failure defined by at least one of the following criteria (5):

  • Age > 65 years
  • A cardiac comorbidity among: systolic dysfunction (ejection fraction ≤45%), permanent atrial fibrillation, or a history of cardiogenic pulmonary edema or myocardial ischemia
  • A respiratory comorbidity among: COPD, restrictive lung disease, or obesity hypoventilation syndrome

• Written informed consent obtained from the patient, or in case of incapacity, from the next of kin/trusted person

• Patient covered by social security

Exclusion Criteria:

• Absolute contraindication to NIV (patient refusal, undrained pneumothorax, intractable vomiting, upper airway obstruction, upper gastrointestinal bleeding, severe craniofacial trauma)
• Absolute contraindication to nasogastric tube placement (upper gastrointestinal bleeding, epistaxis, skull base trauma)
• Persons referred to in Articles L1121-5 to L1121-8 of the CSP (corresponds to all protected persons: pregnant woman, parturient woman, breastfeeding mother, person deprived of liberty by judicial or administrative decision, persons receiving psychiatric care under Articles L. 3212-1 and L. 3213-1 who do not fall under the provisions of Article L. 1121-8, persons admitted to a health or social establishment for purposes other than research, minors, person subject to a legal protection measure or unable to express their consent).