The investigators assessed the efficacy and safety of this noninvasive Ventilation (NIV) bundle strategy compared with a conventional treatment in postoperative patients with ARF.

This is a randomized, prospective, open-label study. Postoperative patients with ARF included in the study were randomly assigned to intervention group or conventional treatment group. In the intervention group, NIV was early used for treatment of postoperative respiratory failure, if patients were inability to spontaneously clear airways from excessive secretions during NIV, fiberoptic bronchoscopy (FBO) was used for suction of secretions , if patients showed intolerance or inadequate patient cooperation during the NIV session, they were sedated (Ramsay scale 2-3) by a continuous perfusion of propofol during the NIV session.

Patients assigned to the control group received supplemental oxygen, Respiratory therapists delivered this intervention using conventional masks or venturi oxygen, and adjusted FiO2 to achieve arterial O2 saturation of more than 92%. The application of noninvasive ventilation was considered, if patients failed the supplemental oxygen and met at least two of the following criteria:(1) severe respiratory distress with dyspnoea, respiratory rate>30breaths/min, and clinical signs suggestive of respiratory-muscle fatigue or increased respiratory effort (use of accessory respiratory muscles or paradoxical abdominal breathing, or intercostal indrawing), (2) respiratory acidosis (arterial pH of 7.35 or less with PaCO2 of 45 mm Hg or more) ;(3) arterial O2 saturation by pulse oximetry less than 90% or PaO2 less than 60 mm Hg at an FiO2> 0.5 or breathing at least 10 l/min oxygen.

All the patients were continuous monitored of vital signs.whereas arterial blood gases were analyzed before NIV and 1h after NIV, before and after FBO, and once a day and in the case of any change either in the ventilatory settings or in the FIO2; duration of noninvasive ventilation, the rate and cause of endotracheal intubation, ICU and hospital mortality and length of stay in the ICU and in hospital, study procedure related complications.

Inclusion Criteria:

• Postoperative patients with acute respiratory failure (ARF) were included in the study,who had to meet one of the following criteria
• Respiratory acidosis (arterial pH of 7.35 or less with PaCO2 of 45 mm Hg or more)
• Partial pressure of arterial oxygen (PaO2)/fraction of inspired oxygen (FiO2) ratio less than 250
• Dyspnoea with respiratory rate > 25 breaths/min or use of accessory respiratory muscles or paradoxical abdominal breathing

Exclusion Criteria:

• Cardiac or respiratory arrest; loss of consciousness
• Facial trauma or surgery or deformity sufficient to preclude mask fitting
• Active upper gastrointestinal bleeding
• Haemodynamic instability or unstable cardiac arrhythmia
• Multiple organ failure