Noninvasive ventilation (NIV) can provide ventilatory support in selected patients with acute respiratory failure, for instance due to acute exacerbation of COPD and acute heart failure. Advantages of noninvasive ventilation compared to invasive mechanical ventilation include absence of complications associated with endotracheal intubation, lower risk of pneumonia, lower level or even absence of sedation and the ability of the patient to verbally communicate. However, in approximately 30% of patients NIV fails and endotracheal intubation is needed to provide optimal ventilatory support. Surprisingly, very few studies have investigated why patients fail on NIV. Clinical observations indicated that agitation, delirium and most importantly asynchrony between patient and ventilator play a role in unsuccessful support with NIV. The upper airways are bypassed during endotracheal intubation. However, with NIV the upper airways may play a role in the efficiency of ventilatory support. In normal breathing the upper airways actively dilate before initiation of inspiratory flow. This is a highly appropriate response as it prevents narrowing of the upper airways during inspiration, which would result in elevated inspiratory resistance. Experiments in newborn lambs have shown that NIV has profound effects on physiology of the upper airways. Positive pressure during inspiration results in constriction of upper airway muscles in the early phase of inspiration. This results in elevated upper airway resistance with lower tidal volume delivered to the lungs. Subsequent studies revealed that reflexes that mediate this response originate in vagal afferences located in the lower airways. From an evolutionary point of view this might be an appropriate response, as high pressure delivered to the lungs may induce barotraumas. However, these responses may negatively affect the efficiency of ventilatory support delivered during NIV. The understanding of upper airway constriction and dilation during NIV is rudimentary. This study aims at determining the effect of NIV on regulation of upper airway patency in patients with COPD.

Inclusion Criteria:

• Informed consent
• COPD
• Hypercapnic respiratory acidosis
• Clinical need of NIV ventilation on the intensive care
• NAVA catheter in situ

Exclusion Criteria:

• Pre-existent muscle disease (congenital or acquired) or diseases / disorders known to be associated with myopathy including auto-immune diseases.
• Diabetes
• Upper airway/esophageal/mouth or face pathology (i.e. recent surgery, esophageal varices, diaphragmatic hernia)
• Recent (< 1 month) nasal bleeding
• Allergic to xylocaïne

The patient population for this study will be included from the intensive care. All patients at the intensive care unit who are in clinical need of noninvasive mechanical ventilation due to hypercapnic COPD and with a NAVA catheter in situ, will be screened and asked for informed consent to participate.