Nasal high-flow therapy and dispersion of nasal aerosols in an experimental setting Sally Roberts, N. Kabaliuk, Cjt Spence, Jane O'Donnell, Z. Zulkhairi Abidin, R. Dougherty, S. Roberts, Y. Jiang, Mc Jermy University of Canterbury, New Zealand Fisher & Paykel Healthcare, New Zealand University of KS, USA Auckland District Health Board, New Zealand University of Auckland, New Zealand Background/Purpose: Nasal high-flow (NHF) therapy delivers flows of heated and humidified gases up to 60 L/min via a nasal cannula. NHF is widely used to support patients, including those who may be infectious. It is thought that NHF gas flow velocities may increase cross-infection risk. Methods: Aerosols within the exhaled breath of healthy volunteers were imaged. Experimental breathing conditions deemed as typical patient breathing conditions were tested: at rest, with a violent exhalation (snorting), both with and without NHF, at flows of 30 and 60 L/min, and for both separate nostrils. The number, diameter, evaporation rates, and velocity of exhaled aerosols were collected. Results: The numbers of aerosols measured were greatest during a violent exhalation without NHF and reduced with NHF. The numbers of aerosols were higher at 60 than 30 L/min, suggesting that higher gas flow rates may be associated with increased aerosol production; however, the numbers were on average 43% and 56% less than without NHF, respectively. During breathing at rest, no differences were imaged between with and without NHF, except at 60 L/min where numbers of aerosols produced were equivalent to 10% of a violent exhalation. Aerosol trajectory and evaporation rates observed both with and without NHF predicted that aerosols between 25 and 250 μm may travel up to 4.4 m and remain airborne for 43 seconds. Conclusions: NHF use does not increase the risk of dispersing infectious aerosols above the risk of typical patient breathing with violent exhalation, which is the worst-case clinical scenario; therefore, standard risk control measures should apply.
Read full abstract