Effects of combined oxygen and surgical masks on inspired fraction of oxygen: relevance to COVID-19-induced respiratory failure

Editor—On March 11, 2020, WHO declared the novel coronavirus disease 2019 (COVID-19) outbreak a global pandemic.¹ Some patients with COVID-19 present with acute hypoxaemic respiratory failure. In such cases, low-flow oxygen therapy through a nasal cannula or an oxygen mask and high-flow nasal cannula oxygen therapy are initially used for supportive oxygen therapy before determining the need for invasive mechanical ventilation.²

Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), the causative agent of COVID-19, can be transmitted by aerosol inhalation or contact with fomites.³ The risk of aerosol transmission has been the largest concern among healthcare workers who care for patients with COVID-19 and acute hypoxaemic respiratory failure. The use of a surgical mask is recommended for patients receiving supportive oxygen therapy.⁴ Although placement of a surgical mask over nasal cannulae is the standard, it remains unclear whether a surgical mask should be placed over or under an oxygen mask. In particular, the effect of the surgical mask position relative to the oxygen mask on fraction of inspired oxygen (F_IO₂) remains to be established. This present study was an experimental trial to evaluate the effect of surgical mask placement over or under an oxygen mask on F_IO₂ according to oxygen flow rate.

The experimental trial was performed using INOflo® DS (Mallinckrodt Pharmaceuticals, Hampton, NJ, USA), which enabled continuous monitoring of F_IO₂ changes in a patient gas sample line that drew air at a flow rate of 230 ml min⁻¹. We measured F_IO₂ around the lips through the mouth under normal breathing conditions in five healthy volunteers in three situations: (1) oxygen mask (EcoLite™; Intersurgical, Wokingham, UK) without a surgical mask (Supplementary Fig. S1a); (2) oxygen mask over a surgical mask (Supplementary Fig. S1b); or (3) oxygen mask under a surgical mask (Supplementary Fig. S1c). Oxygen flow rates were set at 5 L min⁻¹ (low), 7 L min⁻¹ (moderate), or 10 L min⁻¹ (high). We collected data on F_IO₂, which reached a plateau around 1 min after changing to each condition. Written informed consent was obtained from all participants.

Data are expressed as a median (inter-quartile range). All statistical analyses were performed using JMP Pro 14 software (SAS Institute, Cary, NC, USA). For multiple comparisons, the Kruskal–Wallis test followed by the post hoc Steel–Dwass test was used. Differences were considered statistically significant at P<0.05.

The five volunteers, aged 29–44 yr, had a median ventilatory frequency of 15 (13–17) bpm. As shown in Supplementary Table S1 and Figure 1 , F_IO₂ was higher when wearing an oxygen mask under a surgical mask than when wearing an oxygen mask without a surgical mask and when wearing an oxygen mask over a surgical mask, regardless of oxygen flow rate. At a flow rate of 5 L min⁻¹, F_IO₂ was higher when wearing an oxygen mask without a surgical mask than when wearing an oxygen mask over a surgical mask (P=0.031) (Supplementary Table S1 and Fig. 1a).

Fig 1.

The F_IO₂ measured depending on surgical mask position relative to an oxygen mask. Measurements were made for wearing an oxygen mask over a surgical mask, oxygen mask without a surgical mask, or oxygen mask under a surgical mask at oxygen flow rates of (a) 5 L min⁻¹, (b) 7 L min⁻¹, and (c) 10 L min⁻¹. F_IO₂, fraction of inspired oxygen.

This report described two important points. First, at a low oxygen flow rate of 5 L min⁻¹, F_IO₂ decreased when wearing an oxygen mask over a surgical mask compared with wearing an oxygen mask without a surgical mask. However, a surgical mask needs to be placed under an oxygen mask to prevent aerosol transmission to healthcare workers, because it cannot entirely cover an oxygen mask. In this situation, oxygen flow rates have to be adjusted, because the expected F_IO₂ cannot be administered. Second, F_IO₂ increased when wearing an oxygen mask under a surgical mask regardless of the oxygen flow rate. In this situation when an unexpectedly higher F_IO₂ can be administered, patients would need to be carefully observed in order to not miss the appropriate timing of intubation, if necessary.

Our results are in line with those of Montiel and colleagues⁵ who showed that a surgical mask over a high-flow nasal cannula device improved blood oxygen level in patients with COVID-19. These findings could be explained by a decrease in room air entrainment diluting the gas mixture.⁵ In contrast, Binks and colleagues⁶ reported that F_IO₂, which was examined using a carbon dioxide sample line attached to a 16 G cannula, did not differ according to the position of the surgical mask relative to a Hudson oxygen mask at a flow rate of 6 L min⁻¹. However, the figure in their report showed significantly higher expiratory oxygen concentrations when wearing an oxygen mask under a surgical mask than when wearing an oxygen mask over a surgical mask. Because different monitoring methods and conditions might yield inconsistent F_IO₂ measurements, direct measurement of blood oxygen levels rather those of F_IO₂ should be considered in further studies to examine the influence of surgical mask position on oxygenation in patients with acute hypoxaemic respiratory failure.

Our study included a small number of participants and was performed under normal respiratory conditions in healthy participants. When using an oxygen mask, F_IO₂ can change with ventilatory frequency and tidal volume.⁷ Therefore, future studies with a larger number of participants and under altered respiratory conditions, such as different ventilatory frequencies and tidal volumes, are needed to evaluate the effect of surgical mask position on F_IO₂ when using an oxygen mask.

Declarations of interest

SH reports grants from Teijin Pharma, outside the submitted work.

Appendix A. Supplementary data

The following is the Supplementary data to this article: