Evaluation of droplet and aerosol dispersion under high flow nasal cannula with or without surgical mask

ABSTRACT

Rationale Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), which causes coronavirus disease (COVID-19), transmit by droplet and aerosol particles. Droplets and aerosol generation during the oxygen delivery methods such as high flow oxygen therapy (HFNC) and noninvasive positive pressure ventilation (NPPV) during COVID-19 respiratory care, may poses a risk of increasing transmission to healthcare workers. We aimed to evaluate droplet and aerosol dispersion associated with oxygen delivery modes, and further to verify the effect of surgical mask (SM) on preventing particle dispersion.Methods: Two experiments were performed at the laboratory of Shin Nippon Air Technologies, Japan, to visualize (Experiment 1) and to quantify (Experiment 2) dispersing particles. Three (Experiment 1) and five (Experiment 2) healthy Japanese male volunteers aged 30-40s and non-smokers, were recruited. For visualization study (Experiment 1), dispersing particles (>5μm) were recorded by ultra-high sensitive video camera 'eye scope'. For quantification study (Experiment 2), two types of micro-particle detection panel 'Type S' which counts particles > 0.5μm or >5μm were used under air-controlled room with down-flow of 0.3m/sec to avoid contamination of dusts and to drop aerosols on Type S panel. Five patterns of oxygen delivery modalities (No device, 5L/min of nasal cannula, 30L/min or 60L/min of HFNC, 10L/min of oxygen mask, and NPPV) with and without SM, while three breathing patterns (rest breathing, speaking, and coughing) were recorded. The differences in continuous numbers between corresponding two groups were analyzed by ratio paired t-test. A P-value <0.05 was considered as statistically significant.Results: Droplets were able to visualize at further than 50cm while speaking, and further than 1m while coughing. Without SM, droplets were more visible with nasal cannula compared to HFNC. SM effectively reduced droplets under each oxygen delivery modes, and they are hard to visualize even in speaking or coughing. In NPPV mode, floating droplets were visible while coughing. Droplets and aerosols were counted 10-times more while coughing compared to speaking. SM significantly reduced both of droplets and aerosol dispersion while speaking or coughing regardless of oxygen delivery mode. Reduction rate of dispersion under HFNC was higher compared to nasal cannula. 60L/min of HFNC did not increase droplets or aerosol dispersion by counts or by distance compared to 30L/min of HFNC. SM effectively reduced over 90% of droplets and over 95% of aerosols during HFNC mode.Conclusions: SM over HFNC mode may be used safely in appropriate infection control setting and recommended for acute hypoxemic respiratory failure in COVID-19 patients.