Quantitative evaluation of aerosol generation during manual facemask ventilation

Summary Manual facemask ventilation, a core component of elective and emergency airway management, is classified as an aerosol‐generating procedure. This designation is based on one epidemiological study suggesting an association between facemask ventilation and transmission during the SARS‐CoV‐1 outbreak in 2003. There is no direct evidence to indicate whether facemask ventilation is a high‐risk procedure for aerosol generation. We conducted aerosol monitoring during routine facemask ventilation and facemask ventilation with an intentionally generated leak in anaesthetised patients. Recordings were made in ultraclean operating theatres and compared against the aerosol generated by tidal breathing and cough manoeuvres. Respiratory aerosol from tidal breathing in 11 patients was reliably detected above the very low background particle concentrations with median [IQR (range)] particle counts of 191 (77–486 [4–1313]) and 2 (1–5 [0–13]) particles.l‐1, respectively, p = 0.002. The median (IQR [range]) aerosol concentration detected during facemask ventilation without a leak (3 (0–9 [0–43]) particles.l‐1) and with an intentional leak (11 (7–26 [1–62]) particles.l‐1) was 64‐fold (p = 0.001) and 17‐fold (p = 0.002) lower than that of tidal breathing, respectively. Median (IQR [range]) peak particle concentration during facemask ventilation both without a leak (60 (0–60 [0–120]) particles.l‐1) and with a leak (120 (60–180 [60–480]) particles.l‐1) were 20‐fold (p = 0.002) and 10‐fold (0.001) lower than a cough (1260 (800–3242 [100–3682]) particles.l‐1), respectively. This study demonstrates that facemask ventilation, even when performed with an intentional leak, does not generate high levels of bioaerosol. On the basis of this evidence, we argue facemask ventilation should not be considered an aerosol‐generating procedure.


Introduction
The COVID-19 pandemic continues to place unprecedented demands on healthcare globally. The use of airborne personal protective equipment (PPE) has been reserved largely for healthcare workers undertaking medical procedures deemed to be aerosol generating [1][2][3]. These procedures are presumed to generate as much or higher levels of bioaerosols from the respiratory tract than coughing and consequently carry an increased risk of viral transmission. The evidence for these putative 'aerosol-generating procedures' is predominantly epidemiological and from the time of the SARS-CoV-1 epidemic in 2003 [4,5]. Several recent studies have questioned whether these medical procedures should be classified as 'aerosol generating' following quantitation of the aerosol produced during these activities [6][7][8].
Facemask ventilation is a core airway intervention and listed by the World Health Organization (WHO) as "aerosol generating" [4,5]. The epidemiological evidence for this designation is from a single study that reported an increased risk of SARS-CoV-1 infection from facemask ventilation before tracheal intubation [9]. This risk was  [4,9]. Twenty-two of the 26 healthcare workers were infected by one patient and performing an electrocardiogram was associated with an even higher risk of SARS-CoV-1 infection (OR 3.5, 95%CI 1.6-7.9).
No study to date has quantified specifically the aerosol generated during facemask ventilation. Two recent clinical studies, performed in operating theatres, demonstrated relatively little aerosol generation for laryngoscopy and tracheal intubation [6,10]. The analysis of the phase of facemask ventilation of the anaesthetised patient, before tracheal intubation, demonstrated conflicting results. Brown et al. reported that facemask ventilation was not aerosolgenerating [6]. In contrast, Dhillon et al. recorded an increased particle concentration above background during a period including facemask ventilation [10]. Resolution of these different findings is of crucial importance, as facemask ventilation is a key component of elective and emergency airway management. We therefore co-developed an experimental protocol to test specifically whether facemask ventilation is a high-risk procedure for aerosol generation.
To assess the relative risk, we measured aerosol generation during facemask ventilation and compared this against tidal breathing and volitional coughs, with patients as their own controls.

Methods
The study protocol was approved by the Greater Manchester Research Ethics Committee as part of the AERATOR study. The methods for aerosol measurement have been described previously [6]. In brief, a prospective environmental monitoring study was conducted in operating theatres in a UK hospital (Southmead Hospital, North Bristol NHS Trust). All recordings were made within operating theatres with an ultraclean ventilation system (EXFLOW 32, Howorth Air Technology, Farnworth, UK) placed in standby mode [11,12]. This provides an Based on previous work [6], the starting hypothesis was that facemask ventilation would produce no increase in aerosol above background. The sampling methodology was similar to previous work investigating aerosol production during gastro-oesophageal endoscopy [11], where breathing was clearly distinguishable above background. Sample size calculations predicted 10 participants would ensure the study would be adequately powered to detect a difference of clinically important magnitude [13].
Data were processed in the TSI Aerosol Instrument particles.l -1 , which was no different to background (p = 0.43) and much lower than the concentration recorded during tidal breathing (p = 0.001) (Figs. 1 and 2a).
The particle concentration during facemask ventilation with a leak was 11 (7-26 [1-62]) particles.l -1 , which was approximately five-fold higher than background (p = 0.019) 24 but still much lower (17-fold) than that seen during tidal breathing (p = 0.002). The analysis of the difference in particle concentration between facemask ventilation with and without a leak showed no statistically significant difference (p = 0.074) (Fig. 2a).

Discussion
This study demonstrates that facemask ventilation in anaesthetised patients, even with a leak, generates less aerosol than tidal breathing and far less aerosol than a cough. This supports the findings from our previous study which included periods of facemask ventilation as part of  the intubation sequence [6]. We found no evidence that the procedure of facemask ventilation in these circumstances generates high aerosol concentrations and therefore it should not be classified as an aerosol-generating procedure [1,15]. This has implications in a wide range of settings including during routine anaesthetic airway management. The avoidance of facemask ventilation before tracheal intubation or supraglottic airway insertion, due to concerns around aerosol generation, is not supported by this new evidence and likely serves only to increase the risk of encountering difficulties in airway management.
We have used tidal breathing and cough from participants to enable within-subject comparison and relative risk estimation for facemask ventilation. Inter-patient variation was considerable and ranged up to 50-fold for tidal breathing and 36-fold during coughing. This is in keeping with previous studies performed by the AERATOR group and others [6,8,11,12,16]. Therefore, using each participant as their own reference increases the power to generate meaningful comparisons from a relatively small sample. We have also modified our aerosol sampling position to move from 0.5 m to 0.2 m so as to be closer to the mouth of the patient. This has increased our ability to detect the emitted aerosol from source and has increased the concentration of particles recorded with tidal breathing and other respiratory activities. There was a 48fold increase in particle concentration detected during tidal breathing when recorded at this closer position (191 vs. 4 particles.l -1 ) compared with our previous study of supraglottic airway devices performed in the same environment [12]. The higher measured particle concentration closer to the mouth are likely due to decreased particle dispersion and the capture of particles with low momentum in the smaller size range when sampling at 0.2 m compared with 0.5 m. Despite sampling closer to the source, it is possible some aerosol was not detected during facemask ventilation. However, as the concentration detected was far lower than that produced by tidal breathing, we can infer the relative risk of aerosol generation by facemask ventilation is very low.
The low concentration of aerosol detected during facemask ventilation with an intentional leak is also reassuring given that this represents a worst-case scenario.
The particles detected during facemask ventilation with leak likely represent respiratory aerosol originating from the lungs and upper airways during continued positive pressure ventilation rather than from turbulent airflow over the face. This is supported by the fact that these particles were predominantly smallwhich is consistent with respiratory origin where the smallest particles are thought to be generated [17]. We emphasise, however, that this concentration of aerosol was far lower than the patient would generate if conscious and breathing at rest. We can extend this conclusion further to state that a well-fitting facemask with a good seal reduces emitted aerosol concentration to the point where it is indistinguishable from the near-zero aerosol background, and we have previously demonstrated a well-fitting facemask can prevent bioaerosol leak from a cough [12] by keeping respiratory aerosols within the breathing circuit. This is entirely predictable as the mask forms a physical barrier to aerosol spread.
A limitation of our study is that we intentionally studied a period of facemask ventilation after neuromuscular blockade to focus on the aerosol generation associated with the specific procedure rather than any paroxysmal respiratory event like coughing. However, aerosol sampling was conducted continuously throughout the induction of anaesthesia and the period of facemask ventilation performed immediately before the formally analysed period (i.e. before neuromuscular blockade) did not show increased aerosol concentrations above background (Fig. 1). Previous work performed by our group quantified aerosol generation during facemask ventilation in anaesthetised patients without neuromuscular blockade, which again did not demonstrate aerosol generation [12]. We are confident that our conclusions may be generalisable to the unparalysed patient. generating are not intrinsically high risk for generating aerosol, and that natural patient respiratory events often generate far higher levels [6,7]. Furthermore, some of those procedures that generate aerosol, such as gastroesophageal endoscopy, only do so when the patient coughs [11]. The emerging evidence from quantitative clinical aerosol studies is yet to be incorporated into clinical guidance for aerosol generating procedures and we believe this needs urgent reassessment. Declassification of some of these anaesthesiarelated procedures as aerosol generating would seem appropriate due to their lack of aerosol generation. Our findings also raise the broader question of whether the term 'aerosol generating procedure' is still a useful concept for anaesthetic airway management practice in the prevention of SARS-CoV-2 or other airborne pathogens [18].

Acknowledgements
The AERATOR study is registered in the ISRCTN registry