A quantitative evaluation of aerosol generation during awake tracheal intubation

Aerosol‐generating procedures are medical interventions considered high risk for transmission of airborne pathogens. Tracheal intubation of anaesthetised patients is not high risk for aerosol generation; however, patients often perform respiratory manoeuvres during awake tracheal intubation which may generate aerosol. To assess the risk, we undertook aerosol monitoring during a series of awake tracheal intubations and nasendoscopies in healthy participants. Sampling was undertaken within an ultraclean operating theatre. Procedures were performed and received by 12 anaesthetic trainees. The upper airway was topically anaesthetised with lidocaine and participants were not sedated. An optical particle sizer continuously sampled aerosol. Passage of the bronchoscope through the vocal cords generated similar peak median (IQR [range]) aerosol concentrations to coughing, 1020 (645–1245 [120–48,948]) vs. 1460 (390–2506 [40–12,280]) particles.l-1 respectively, p = 0.266. Coughs evoked when lidocaine was sprayed on the vocal cords generated 91,700 (41,907–166,774 [390–557,817]) particles.l-1 which was significantly greater than volitional coughs (p < 0.001). For 38 nasendoscopies in 12 participants, the aerosol concentrations were relatively low, 180 (120–525 [0–9552]) particles.l-1, however, five nasendoscopies generated peak aerosol concentrations greater than a volitional cough. Awake tracheal intubation and nasendoscopy can generate high concentrations of respiratory aerosol. Specific risks are associated with lidocaine spray of the larynx, instrumentation of the vocal cords, procedural coughing and deep breaths. Given the proximity of practitioners to patient‐generated aerosol, airborne infection control precautions are appropriate when undertaking awake upper airway endoscopy (including awake tracheal intubation, nasendoscopy and bronchoscopy) if respirable pathogens cannot be confidently excluded.


Introduction
Airborne transmission of SARS-CoV-2 in fine aerosols is now widely recognised [1,2]. The concept of aerosol-generating procedures (AGP) emerged following the SARS outbreak in which healthcare practitioners acquired nosocomial SARS-CoV-1 [3,4]. Most healthcare agencies have not changed the list of medical procedures classified as AGPs since the start of the pandemic, but emergent clinical evidence has shown many medical interventions and procedures do not inherently generate aerosols [5,6]. A recent UK Infection Prevention and Control review resulted in the removal of several procedures from the AGP list in England [7][8][9][10][11][12].
These clinical aerosol monitoring studies complement existing knowledge that humans generate bio-aerosol during normal respiratory events such as breathing, speaking, coughing and sneezing which facilitate disease transmission [13][14][15][16].
Studies performed in anaesthetised patients have demonstrated tracheal intubation is not high risk for aerosol generation; however, awake tracheal intubation (ATI) is performed in awake or lightly sedated patients who can breathe, speak and cough [7,17]. In this respect, ATI may be similar to oesophagoduodenoscopy, which is associated with a high risk of aerosol generation when performed in the awake or sedated patient and has therefore been retained on the list of AGPs in England [18].
Although ATI is undertaken for < 1% of general anaesthetics [19], it remains a core technique for airway management in patients with a predicted difficult airway [20][21][22]. Tracheal intubation undertaken with a flexible bronchoscope in an awake or sedated patient shares many similarities with flexible bronchoscopy and nasendoscopy.
A small study conducted early in the pandemic suggested awake bronchoscopy did not increase aerosol generation at group level; however, 10% of bronchoscopies were associated with increased aerosol concentrations. Caution was recommended regarding its designation as an AGP [23].
To better assess the risk of aerosol generation during ATI and nasendoscopy, we undertook a clinical monitoring study during an ATI training course. The aerosol generated during the procedure was compared with that generated during each participant's natural respiratory activities and background concentrations.

Methods
We conducted a prospective study in a UK hospital (University Hospitals Bristol and Weston NHS Trust, Bristol, UK) during an ATI training course. Ethical approval was granted by the Greater Manchester Research Ethics Committee. All study participants provided written informed consent. The study was undertaken in an operating theatre with an ultraclean ventilation system (EXFLOW 90, Howorth Air Technology, Farnworth, UK) placed in standby mode. This results in a very low background aerosol concentration with an air exchange rate equivalent to standard UK operating theatres as described previously [18,24].
All participants were anaesthetists in training who both performed and underwent an ATI (see online Supporting Information Appendix S1). The research team were not involved in the course teaching. An optical particle sizer (Model 3330, TSI Incorporated, Shoreview, MN, USA) sampled aerosol (0.3-10 lm diameter, at 1 Hz, flow rate of 1 l.min -1 ) connected to a 3D-printed funnel as described previously [17]. Bronchoscopy was undertaken with an Olympus OTV-SI camera system and LF-V scopes (4.1 mm external diameter, Olympus, Southend-on-Sea, UK).
Baseline aerosol sampling was performed before ATI, Following tracheal intubation, the tracheal tube was connected to the in-circuit aerosol sampling set-up (Fig. 1b) to enable recording of exhaled aerosol from the awake, intubated participant for 60 s. If tolerated, the tracheal tube was then withdrawn to 16 cm at the nares (to move the distal tip above the vocal cords) for up to a further 60 s of in-circuit aerosol sampling. Following tracheal extubation, the sampling funnel was repositioned 20 cm in front of the The link between aerosol emission rate and risk of respiratory pathogen transmission remains unknown. We considered a 2-fold increase in the aerosol concentration measured during the procedure, above baseline tidal breathing, to be clinically relevant. Using data from our previous facemask ventilation study [12], a non-parametric paired comparison predicted that recordings of complete data from 12 participants would ensure the study was adequately powered to detect a difference of this magnitude (80% power, a 0.05, G*Power3.1.9.4). We used TSI Aerosol Instrument Manager software to process aerosol data before analysis in R (R Foundation, Vienna, Austria) and Origin Pro (OriginLab, Northampton, MA, USA). Comparisons were made between aerosol measurements using Wilcoxon matched pairs tests.

Results
The study was performed on a single day during an ATI course. All 12 participants (6 female) were specialist anaesthetic trainees who both performed and received ATI.   (Fig. 9).

Discussion
We have shown that ATI without sedation generates high their study was likely underpowered to draw definitive conclusions [23,25]. Additionally, we have been able to examine aerosol generation during nasendoscopy and believe it is the first study to examine this procedure in an ultraclean environment. Over a fifth of nasendoscopies were associated with an obvious respiratory event such as a periprocedural cough, sneeze or deep breath, half of which had an associated large aerosol peak. Four nasendoscopies without obvious expulsive respiratory activities were also associated with high peak aerosol concentrations. Our findings extend previous studies, performed in nonultraclean environments, which demonstrate that breathing, coughing, sneezing and phonation during nasendoscopy generate respiratory aerosol [26][27][28]. These studies had conflicting conclusions as to whether nasendoscopy/ laryngoscopy should be defined as an AGP. However, as the procedure is commonly and somewhat unpredictably associated with levels of aerosol generation that are comparable to volitional coughs, directed towards the operator who will be near the patient, it is logical to Respiratory aerosol is thought to be generated from three main sources: the oropharynx, which is considered to generate larger droplets; the laryngeal mode, which generates aerosols in the 1-2 lm range as air passes over the vocal cords and becomes more turbulent; and the alveolar mode, which generates smaller particles due to thè fluid-film burst´as collapsed alveoli re-expand on inspiration [16]. The aerosol concentrations recorded during tidal breathing were lower than those detected in our other studies [12,17,18,24]. It has been reported previously that young healthy participants do not produce significant aerosol during tidal breathing [16]; however, when sampling within the circuit, all participants generated measurable aerosol (Fig. 7). As such, there is a reduction in aerosol detection when sampling at 20 cm distance from the patient's mouth due to particle dilution and dispersion.
All participants were healthy anaesthetists in training which may not reflect the patient population who have ATIs; however, as we found this procedure to generate high concentrations of aerosol, we feel this will have provided a realistic lower estimate of the risk. It is likely that patients with other comorbidities, including respiratory conditions, may generate higher aerosol concentrations. The sample size is relatively small but having the ability to make withinsubject comparisons in this dataset increases the power to make statistical inferences on the findings. We have only studied awake participants and our results may not be transferrable to sedated patients. The Difficult Airway Society ATI guidelines advise cautious use of minimal sedation as it may improve procedural success, however, the guidelines emphasise sedation is neither essential nor a substitute for inadequate airway topicalisation [22]. Our study demonstrated 100% intubation success without additional sedation.
Endoscopist-administered midazolam sedation did not decrease the likelihood of coughing during upper gastrooesophageal duodenoscopy [18]; however, opioid and other sedatives with an antitussive effect may reduce the frequency and extent of coughing; whether these sedatives will sufficiently reduce the risk of aerosol production to a safe level is unknown.
We were unable to measure aerosol during airway topicalisation as nebulised lidocaine generates very high aerosol concentrations which would mask any generated aerosol. All participants were anaesthetists-in-training who were aware of the broad scope of the study; however, the participants and operators were unable to view particle analysis data at any point during the intubations and nasendoscopies, and the research team had no influence on the conduct of the procedures. Awake tracheal intubation can also be performed using videolaryngoscopy and though not studied, this technique is likely to have comparable results as the invasive elements of the procedures are similar and the patient remains able to breathe, cough and speak.
Our findings contrast with the reassuring findings of our previous AGP studies which demonstrate tracheal intubation of anaesthetised patients is not high risk for generating aerosol [7,17]. Awake tracheal intubation generates 3-4 orders of magnitude more aerosol than intubation of anaesthetised patientsthe obvious difference between these studies is the conscious state of the subject [7,17]. When high aerosol concentrations are generated during ATI, this is most often due to natural respiratory events such as breathing, speaking and coughing. Therefore, we feel that it is inappropriate to place Figure 6 Average and peak aerosol concentration for all nasendoscopies. Black line represents median value. Grey circle, no event; red triangle, cough; orange square, big breath; crosshair, gag; blue cross in a square, vocalise; pink star, sneeze.

Figure 7
In-circuit aerosol sampling during tidal breathing of a participant following tracheal intubation for 60 s with the distal tip of the tracheal tube (TT) below the vocal cords, followed by 60 s of sampling with the tracheal tube repositioned to ensure the distal tip was above the vocal cords (tube 16 cm at the nares).

Figure 8
In-circuit integrated particle number concentration for breathing by participants. Wilcoxon matched pairs. n = 12 for breathing through tracheal tube held in the mouth and intubated breathing, n = 9 for breathing with distal tip above the cords. ns, p > 0.05; *, p < 0.05, **, p < 0.01. the emphasis of risk assessment for aerosolised respiratory pathogen transmission on the procedure. Our results again question the utility of the overarching term`aerosol generating procedure´ [29].
Our findings demonstrate that performing ATI via a nasally inserted flexible bronchoscope in a conscious participant consistently generates a high concentration of respiratory aerosol. Airborne level personal protective equipment is therefore required when this procedure is undertaken in a patient in whom respiratory pathogens cannot be excluded. These findings will help inform guidance for management of patients undergoing ATI in operating theatres but are also of relevance to clinicians undertaking bronchoscopy and nasendoscopy.

Acknowledgements
This study was part of the NIHR-funded AERATOR study. AS is

Figure 9
In-circuit size particle distribution analysis. Dot, median value; error bars, IQR. Distal tip of tracheal tube in pharynx (n = 9); below vocal cords and in mouth pre-intubation. Black line, pharynx; grey line, below vocal cords; yellow line, oral cavity.