3.1. Effects of masks on carbon dioxide re-breathing
In the study of Ulrike Butz's dissertation [18] focusing on possible rebreathing of carbon dioxide in 15 healthy adult male volunteers, a carbon dioxide partial pressure of up to 21-24 mmHg was found under a surgical mask after 30 minutes [18]. This corresponds to about 2.8 - 3.2 % carbon dioxide of the inhaled air under the mask.
In Pifarrés mask-experiments in 8 adult females and males a health-critical value of carbon dioxide concentration (CO2 Vol%) was measured in the air under the masks after few minutes. The concentrations of 14162 ppm with a mask versus 464 ppm without a mask were statistically significant with p <0.001 increased by a high factor compared to the initial value (ambient air) and even more following exercise [19]. According to those experiments, masks can be responsible for a drastic increased CO2 concentration of the inhaled air, which roughly corresponds to 1.41-1.7% carbon dioxide in inhaled air under the face mask (p < 0.001) [19].
A project at the University of Delft used a validated method that clearly demonstrated that carbon dioxide re-breathing under standardised laboratory conditions (test suite) after 1 minute is at least 0.9% CO2 for N95/ FFP2 masks [20]. Those elevated carbon dioxide levels of inhaled air, particularly under N95 masks, have also been found in physiologic relevant short-time modeling studies. This confirms a constant increase leading to an averaged 1% inhaled CO2 per breath during simulations of eight breathing cycles in 33.65 seconds [21] (see Figure 8 of mentioned publication with animation of CO2 distribution with and without a respirator). Another modeling study shows that wearing N95 masks results in carbon dioxide accumulation, the volume fraction of CO2 reaches 1.2% after 7 breathing cycles and is then maintained at 3.04% on average. The wearers re-inhale excessive CO2 with every breath taken from the mask cavity [22].
In 2012 Sinkule already evaluated 30 different N95 respirators using the NIOSH Automated Breathing and Metabolic Simulator (ABMS) through 5 minute work rates and found elevated CO2 levels in the inhaled air ranging between 1.28% and 3.52% [23]. These results are consistent with measurements of CO2 in the dead space of the masks from experimental studies in humans with values of 2.8 [24] and 3.2 % [25].
In a self-experiment in 2020 Geiss measured the air under masks under laboratory conditions and only found an accumulation of carbon dioxide between 0.22 and 0.29% within 5 minutes mainly under surgical masks [26]. However, this experiment has several limitations. Firstly, it is only a one-time measure performed by a man, which might not be representative. The anatomy of this volunteer does not reflect children or women. Secondly, the CO2 sampling point chosen by Geiss above the tip of the nose on the bridge of the nose is suboptimal for mask measurements. This is because it is not close enough to the openings involved in breathing, which are shielded from the rest of the dead space of the mask by the protruding tip of the nose (see figure 1A in Geiss publication to illustrate the questionable placement of sensor [26] and compare it to the gas distribution video in Salati [21]). Thirdly, it is not optimal to place the sampling point at the highest point. This is because carbon dioxide is heavier than other air components (approximately 44 g/mol in CO2 compared to 32 g/mol in O2 and 28 g/mol in N2) and could accumulate there over time to a lesser extent than in the lower parts of the dead space of the mask [21]. In contrast, Butz provided a positioning of the sensor close to the mouth attached to the cheek [18], like Blad [20] and Sinkule [23], who placed it close to the breathing orifice (mouth opening), while Rhee and Roberge chose the nasolabial fold [25,27].
In a prospective observational study in 2021, Rhee examined the carbon dioxide concentration of 11 healthy volunteers during regular breathing and sitting at rest while they put on different types of masks for 15 minutes. Serial CO2 measurements were performed with a nasal cannula at a frequency of 1 Hz [27]. The measured 2.4-2.6% CO2 concentration translates into a highly significant increase in CO2 with a KN95 respirator and a valved respirator at the nasolabial fold (p < 0.0001), which is much greater than the NIOSH 8h threshold limit value [17]. The National Institute for Occupational Safety and Health (NIOSH) has an 8h threshold limit value – time-weighted average recommended exposure limit (TLV-REL) of 0.5% – and a 15 min threshold limit value – short-term exposure limit (TLV-STEL) of 3% for CO2 – in workplace ambient air [17]. Rhee´s well designed reliable high quality study demonstrates a significant increase in end-tidal CO2 concentrations among healthy volunteers while donning KN95 respirators. Consequently, the authors recommended further studies.
Table 1 summarizes the experimental findings concerning CO2-re-breathing under face masks.