Background: We examined whether SARS-COV-2 virus RNA could be detected in air and environmental samples in the hospital isolation rooms in patients who received a) High flow nasal cannula (HFNC), b) non-invasive ventilation (NIV) and c) conventional oxygen therapy (COT) via nasal cannula for respiratory failure.

Method: A field test at the Prince of Wales hospital isolation room with 12 air changes/hr on patients with COVID-19 confirmed by RT-PCR with nasopharyngeal flocked swabs and throat swabs who required a) HFNC up to 60L/min (n=6), b) NIV (n=6), and c) COT up to 5L/min of oxygen (n=14). Three air samplers were placed around each patient and sampled air for at least 2 hours continuously. Surface samples were collected from all available surfaces after air sampling for 16 patients.

Results: Altogether 3/63, 1/54, and 4/150 air samples were positive from patients receiving HFNC, NIV, and COT, p=0.614, with mean(SD) RT-PCR cycle threshold(CT) values of 36.91(0.85), 38.53, and 37.74(0.71), respectively, p=0.234. In contrast, 3/44, 14/72, and 8/59 environmental samples were positive in patients receiving HFNC, NIV, and COT, p=0.166, with CT values 35.96 (0.96), 36.47 (2.34), and 35.61 (2.53), respectively, p=0.699. Lower respiratory specimen CT values on the sampling day were associated with positive air samples [OR 0.83 (95%CI 0.70 ? 0.99), p=0.034].

Conclusion: HFNC and NIV did not increase the risk of air or environmental contamination compared to COT in patients with respiratory failure due to COVID-19. A high viral load on the day of air sampling was associated with positive air samples. (supported by HMRF#COVID190110)