Current evidence suggests a mechanistic and physiological rationale for the use of high flow nasal cannula (HFNC) in acute respiratory hypoxemic failure (AHRF) based on physiological studies in airway models, healthy volunteers and patients with Chronic Obstructive Respiratory Disease (COPD). This is supported by observational studies in patients with AHRF with reductions in a range of respiratory and other physiological parameters. Observational studies also suggest similar intubation rates and lower failure rates with HFNC when compared to non-invasive ventilation (NIV) with improved patient acceptance and tolerance for HFNC.

The role of HFNC is less clear in acute hypercapnic respiratory failure. Although non-invasive ventilation is the recommended treatment, it is associated with discomfort, and a significant proportion (up to 25% in some reports) cannot tolerate non-invasive ventilation. Observational reports and limited data from randomized controlled trials suggests that HFNC is effective in treating patients with hypercapnic respiratory failure.

We designed this trial to assess whether early application of HFNC in patients with non-severe hypercapnic respiratory failure can correct barometric abnormalities, and prevent progression to non-invasive ventilation or tracheal intubation and mechanical ventilation.

Inclusion Criteria:

• Adult patients > 18 years of age
• Acute Hypercapnic respiratory failure with pH < 7.35 and pCO2 > 45 mmHg

Exclusion Criteria:

• Pregnant or Breast-Feeding
• Patients who cannot read and understand French or English
• Hypercapnia secondary to a drug toxicity or non-pulmonary aetiology
• Hypercapnia secondary to exacerbation of asthma
• Contraindication to NIV
• Contraindication to HFNC
• Not for escalation to NIV based on a ceiling of care
• pH < 7.15
• GCS 8 or less
• Shock defined as systolic < 90 mmHg or a reduction by 20mmHg from usual systolic BP despite volume resuscitation
• Respiratory or cardio-respiratory arrest
• Any other indication that requires immediate invasive/non-invasive mechanical ventilation

Data may be shared to researchers who will request it, based on reasonable research objectives, ethical approval if needed, and signing of a data sharing agreement by the relevant institutions.

Due to the nature of the intervention, nor the participants, nor the treating physicians, nor the investigator can be blinded. However, the outcomes will be analyzed an independent statistician blinded to the treatment assignments.