Does a surgical mask improve oxygenation in COVID-19 patients?

Abstract

To the Editor, For anesthesiologists and operating room nurses, protection from coronavirus disease 2019 (COVID19) infection while on duty is currently a top priority. According to the recommendation issued by the Japanese Society of Anesthesiology in March 2020, following extubation at the end of general anesthesia, the patient should receive oxygen via an oxygen mask placed over a surgical mask [1]. The rationale for this recommendation is the inhibitory effect of the surgical mask on the spread of respiratory aerosols. Binks and colleagues recently investigated whether the fraction of inspired oxygen (FiO2) was different when wearing surgical masks above and below the oxygen mask. They suggested that oxygen supply via the oxygen mask was not disturbed by simultaneous use of a surgical mask [2]. From a slightly different point of view, however, COVID-19 patients with severe hypoxia might require oxygen administration for a long time after extubation. We devised a method of delivering oxygen through a nasal cannula under a surgical mask to obtain maximum oxygenation while preventing aerosol dispersal (Fig. 1). We compared our method with the method recommended by Binks et al. in a patient with COVID-19 being treated at our intensive care unit. The results are shown in Table 1. Partial pressure of oxygen in arterial blood (PaO2) clearly decreased when changing from the nasal cannula to the oxygen mask, recovering once again when returning to the nasal cannula. In contrast, other indices, such as pH, partial pressure of carbon dioxide in arterial blood (PaCO2), base excess, and respiratory rate, remained almost unchanged. These results suggest that oxygen delivered via a nasal cannula worn under a surgical mask might prevent the spread of infection while simultaneously allowing maintenance of a high PaO2 in patients. To verify why our method was advantageous for oxygenation, we decided to estimate inhaled oxygen concentration when oxygen was administered to a healthy subject. Due to technical difficulties, we compared oxygen levels in exhaled air rather than inhaled air: Exhaled oxygen concentration was 23% with the oxygen mask, and 33% with the nasal cannula (Supplemental Figure 1). Considering that oxygen consumption during these measurements was constant,