Aerosol emission from the respiratory tract: an analysis of aerosol generation from oxygen delivery systems

Introduction continuous positive airway pressure (CPAP) and high-flow nasal oxygen (HFNO) provide enhanced oxygen delivery and respiratory support for patients with severe COVID-19. CPAP and HFNO are currently designated as aerosol-generating procedures despite limited high-quality experimental data. We aimed to characterise aerosol emission from HFNO and CPAP and compare with breathing, speaking and coughing. Materials and methods Healthy volunteers were recruited to breathe, speak and cough in ultra-clean, laminar flow theatres followed by using CPAP and HFNO. Aerosol emission was measured using two discrete methodologies, simultaneously. Hospitalised patients with COVID-19 had cough recorded using the same methodology on the infectious diseases ward. Results In healthy volunteers (n=25 subjects; 531 measures), CPAP (with exhalation port filter) produced less aerosol than breathing, speaking and coughing (even with large >50 L/min face mask leaks). Coughing was associated with the highest aerosol emissions of any recorded activity. HFNO was associated with aerosol emission, however, this was from the machine. Generated particles were small (<1 µm), passing from the machine through the patient and to the detector without coalescence with respiratory aerosol, thereby unlikely to carry viral particles. More aerosol was generated in cough from patients with COVID-19 (n=8) than volunteers. Conclusions In healthy volunteers, standard non-humidified CPAP is associated with less aerosol emission than breathing, speaking or coughing. Aerosol emission from the respiratory tract does not appear to be increased by HFNO. Although direct comparisons are complex, cough appears to be the main aerosol-generating risk out of all measured activities.


Study design
This was performed as part of the wider AERATOR study to assess the risk of aerosolised transmission of SARS-CoV-2 in healthcare settings. Ethical approval was given by the North West Research Ethics Committee (Ref: 20/NW/0393, HRA Approved 18/9/20).

Aerosol measurement
Aerosol measurements were recorded using two devices simultaneously: an Optical Particle Sizer (OPS) and Aerodynamic Particle Sizer (APS). Technical specifications were detailed in a previous publication from our group but are replicated here. 1 The key difference was the use of a shorter length of sampling tube (0.45m) from the sampling funnel through to the instrument inlet as patients were seated for this experiment, not supine. The OPS (TSI Incorporated, model 3330, Shoreview, NM, USA) samples air at 1 L min −1 and detects particles by laser optical scattering. The OPS reports the particle number concentration and optical size distribution within the diameter range 300 nm to 10 µm with a time resolution of 1 second. The OPS is widely used for aerosol studies from laboratories / clean rooms to more demanding outdoor environments. It is calibrated by the manufacturer using polystyrene latex spheres and its performance conforms to the ISO standard 21501-4:2018. The reported optical size of the particles is based on an assumed refractive index of pure water at 600 nm wavelength (1.333).
Both the APS and OPS were connected to the same sampling funnel, which was 3D printed (RAISE3D Pro2 Printer, 3DGBIRE, Chorley, UK) from PLA with a maximum diameter of 150 mm, cone height of 90 mm with a 10-mm exit port. Two conductive silicone sampling tubes of 0.3 m length and internal diameter 4.8 mm (3001788, TSI) were connected to the neck of the sampling funnel, with one connected to the APS and the other to the OPS.
For baseline measurements and HFNO, the funnel was placed such that the top of the funnel cone was 1 cm from the participant's forehead, and the sampling apex of the funnel was 10 cm from the to have a second measurement on a different date to ensure replicability. For CPAP, sampling was at the filtered exit port of the facemask, and at the point of maximum leak from the mask (measured by an experienced operator). If the maximum leak was less than 50 L/min, a leak was generated by the operator and measurement performed as close as possible to this leak. As with baseline measurements, speaking, breathing, and coughing were recorded during CPAP. Finally, measurements were made as the mask was removed for a small number of participants. CPAP settings were set to our hospital standard (15cm H2O pressure), after initial scoping measurements and earlier research found no difference in aerosol emission with changing pressure, oxygen and humidity settings.
For HFNO, As with CPAP, measurements were made during tidal breathing, speaking, and coughing.
For some participants, we tested the effect of wearing a surgical facemask over the HFNO nasal cannula on aerosol emission. As described in detail in supplement (Supplement S2), we also performed a set of measurements and experiments with four separate HFNO machines in order to identify the source of aerosol that we recorded during our study.
Hospitalised patients with COVID-19 were recruited by study members and had simple baseline measurements performed (e.g. speaking, breathing, and coughing, both with and without surgical facemasks). However, as background aerosol concentration was too high to reliably report aerosol emission from breathing and speaking, we only report aerosol emission from coughing.

Statistical analysis
Aerosol generation differs greatly among people, with an approximate log-normal distribution in number concentration. 3,4 As such, our analysis focussed on comparing the relative aerosol number concentrations from different procedures performed by each individual. We report the number concentration, an intensive property that does not depend on scale (i.e. is independent of the time or volume sampled) as reported by the instruments measured over a sample period, selected to be 1 s. We have reported one of two parameters for each activity: either the peak particle number concentration reported across the full number of samples of the measurement for single, forced exhalations such as coughing (cm -3 ); or, the mean particle number concentration reported as the  Because coughs tend to decrease in strength, for some participants, we did a cough wearing a FRSM first in some participants. Note 2: Speaking was asking patients to count from 1-100, at a set cadence and aiming for a similar volume between participants.
Images of the set up: Figure S1: Line drawing of the set up in theatres from above. A: is the volunteer, B is the measuring equipment table, and C represents the limits of the laminar flow canopy.