Patients affected with severe parenchymal pulmonary diseases, such as Chronic Obstructive Pulmonary Disease (COPD ), may experience dyspnea at rest due to increased work of breathing and reduced oxygenation. The delivery of high-flow humidified nasal oxygen (HFNC) has been shown to have a positive-end-expiratory pressure (PEEP) effect and is able to flush out CO2 from the upper airways, reducing dead space ventilation. Furthermore it has been proven to reduce the respiratory rate shortly after its initiation. These multiple actions offer the potential of changing the respiratory pattern and reducing work of breathing, improving the efficiency of breathing.

In this short-term, physiological, open, randomized, cross-over pilot study the investigator swill describe the effects of varying settings of high-flow nasal oxygen on respiratory rate, tidal volume, and diaphragmatic work of breathing in patients with severe COPD. The investigators will also describe changes in gas exchange and effects on the subjects' comfort and dyspnea and the breathing responses to varying setting of CPAP in the subject population.

HFNC has been shown to have many advantages in the treatment of acutely hypoxemic patients, improving their clinical outcome. The exact mechanism underlying this beneficial effect is still not completely understood. Few studies have analyzed the effect of HFNC on ventilatory pattern and work of breathing. The majority of these studies have focused on the effects in healthy volunteers. Only one study from Braunlich et al. studied the effects of HFNC on COPD and interstitial lung disease (ILD) patients, showing that high-flow nasal oxygen reduces respiratory rate and increases the tidal volume in these patients.

In adults, a low flow range from 5 to 10 L/min is comparable to flow received by standard oxygen devices (nasal cannula or facial mask). Patients with underlying pulmonary diseases, as in our study population, have a higher inspiratory flow demands range (from 30 to 120 L/min during an acute respiratory failure episode) compared to healthy subjects.

We expect to observe physiological changes in our outcomes with the proposed Optiflow ™ settings of a minimal therapeutic flow of 30 L/min, intermediate of 45 L/min, and the maximal flow rate of 60 L/min. There is an extensive clinical experience using high flow rates in these ranges and they are generally very well tolerated. As mentioned above, HFNC generates a Positive End Expiratory Pressure (PEEP) comparable to CPAP range of 4 - 8 cmH2O (the minimal and the maximal PEEP generated by the HFNC).

Future studies, based on this pilot study, will differ from previous ones in the following ways:

1. We are testing a different technology. The Optiflow delivers substantially higher flow rates than in the previous Braunlich study13.That study used a single flow rate of 24 L/min whereas we are examining a range of flows that extend considerably higher (30 to 60L/min). We are interested in determining how the effects of higher flow rates compare to those in the range used in the Braunlich study, but we are not able to compare the devices directly because the latter device is not available in the US. It is important to understand whether there is any efficacy advantage to using the higher flow rates available with the Optiflow.

2. Future studies will aim to understand mechanisms of the effect of high flow nasal oxygen.

   1. Are the effects that we anticipate seeing related to changes in inspiratory muscle effort as determined by measurement of transdiaphragmatic pressure and calculation of the pressure time product of the diaphragm?
   2. Or does the flushing of dead space in the nasopharynx improve ventilatory efficiency so that gas exchange can remain stable or even improve (as determined by measurements of minute volume and transcutaneous PCO2 (PtcCO2)? This has implications for use of HFNC to treat patients with COPD exacerbations who are developing respiratory muscle fatigue.

1) Our focus will be on COPD patients for whom the use of HFNC has not been studied much to date. Most studies have focused on patients with hypoxemic respiratory failure. It is important to understand how HFNC affects breathing pattern and gas exchange in COPD patients because earlier reports suggest that excessive concentrations of oxygen administered to COPD patients retaining CO2 can actually worsen the CO2 retention by blunting respiratory drive. The reduction in respiratory rate and minute volume noted by Braunlich et al could represent a blunting effect of O2 on drive to breathe and could promote greater CO2 retention. By monitoring PCO2, something the Braunlich study didn't do, we can assess this possibility.

2) We wish to evaluate the effect of CPAP on the same breathing indices as with HFNC in our COPD patients. We plan to use the CPAP response as a "positive control", to determine if our population responds as described by CPAP studies in the literature. Prior studies have demonstrated that in patients with severe COPD, using CPAP in the range we are proposing, lowers the diaphragmatic work of breathing and we wish to determine if our population manifests a similar effect.

Thus future studies, based on the data obtained from this pilot study, will extend the Braunlich et al study by evaluating the effects of higher flow rates using a different technology available in the US, determining effects on inspiratory muscle effort, and monitoring gas exchange which is important from both mechanistic and safety perspectives. We hypothesize that the higher flow rates will have a greater blunting effect on breathing pattern than a low flow rate and that there will be an improvement in ventilator efficiency that will be associated with decreased breathing work of the diaphragm.

Inclusion Criteria:

• Subjects are 18 or more years of age
• Chronic respiratory failure, defined as indication for long-term oxygen therapy
• Underlying diagnosis of severe COPD (GOLD stage III or IV)

Exclusion Criteria:

• Recent (<1 month) exacerbation Acute exacerbation is defined as a sudden worsening of COPD symptoms (shortness of breath, quantity and color of phlegm) requiring a change in the baseline therapy.
• Respiratory rate at rest >28/min
• Subject requires > 6 L/min nasal O2 to maintain SpO2 >88% at rest
• Subject has severe dyspnea at rest
• Subject has swallowing disorder or chronic aspiration
• Prior esophageal surgery, known esophageal stricture or any other condition that would place the subject at risk during balloon placement
• Recent (< 1 month) abdominal and thoracic surgery
• Severe coagulopathy (defined as platelet count <5000/μL or international normalised ratio >4)
• Subject is too cognitively impaired to give subjective ratings for visual analogue scale.The PI and the Co-Investigators will assess the patient cognition using the Mini Mental State Examination (MMSE)
• Allergy or sensitivity to lidocaine
• Inability to obtain informed consent
• Pregnancy and breastfeeding