Measurement of airborne particle exposure during simulated tracheal intubation using various proposed aerosol containment devices during the COVID‐19 pandemic

We read with great interest the article by Simpson et al. [1]. We are grateful to the authors for providing the first scientific evaluation on the impact of improvised barrier 'devices' on dispersion of exhaled aerosols. As the authors point out, with the exception of the sealed box with suction, all were found to cause no significant decrease in ambient aerosol particles. In the case of the 'aerosol box', rather worryingly, increases in particle counts were recorded. These findings are highly concerning given the seemingly widespread use of such devices during aerosol-generating procedures in patients with COVID-19. As we reconcile the duty to care for patients in the face of this highly transmissible and potentially deadly disease, the requirement for healthcare worker protection is an issue of respecting basic human rights as much as their psychological need to reduce anxiety in a highly stressful environment. However human and understandable the fear may be, we cannot forget that our work is first based on science. We can, and should, turn to science for potential solutions. Based on their findings, we heartily support the decision by Simpson et al. to remove passive barriers from their intubation protocols. We do, however, feel compelled to comment on several aspects of the study. Excluding the emitted aerosols from ventilation in the room will result in a highly concentrated plume. This will favour aerosol escape via apertures on the rigid box at times of sudden increases in internal pressure. This phenomenon in the Simpson et al. experiments was likely facilitated by the conditions of negative pressure room ventilation [1]. Here, it becomes necessary to think about what is happening outside the box; airflow in the space around the box becomes crucial to understanding the pathways of dispersion and areas of the greatest exposure. The authors positioned the particle counter based on the presumed relevance to laryngoscopists' exposure and recorded the changes in aerosol counts. However, there may have been nearby locations with even higher counts, including those relevant to the assistant, as was alluded to in the manuscript. Conversely, a particle counter located in another position might not have picked up any spikes at all. To avoid trial and error in selecting optimal sampling locations, computational fluid dynamic modelling simulation of the particular room with the appropriately set boundary conditions can be performed [2]. A relatively minor change in position of the particle counter with respect to the room airflow patterns could lead to a significant degree of variation in aerosol dispersion and particle counts. Controlling for the variability of airflow, in addition to humidity and room temperature, may be difficult but is necessary to achieve results that truly reflect the effect of the barrier. We wonder how different the results would be if the experiments were done in a 'positive pressure room', a type of ventilation that is present in most operating theatres. In addition, the position of the laryngoscopist (and the point of origin and direction of flow of their exhaled breath) is dictated by the ergonomic properties of each device. Given that participants wore simple procedure masks, unaccounted for variations in airflow and possible droplet contamination due to their breathing were likely introduced. The finding that the five micron particles were more likely to be sampled outside the aerosol box and the sealed box was puzzling. At the high end of the size spectrum for aerosols, these relatively large particles are likely to settle rapidly. Depending on the exact conditions during baseline measurements, the location of sampling and possible environmental contamination may have contributed to this finding. Lastly, the most interesting result from the problem-solving perspective is related to the performance of the sealed box with suction. By virtue of its construction, the box precluded airway management but performed remarkably well in reducing aerosol egress. To find solutions that truly improve safety, we need sound engineering solutions that are acceptable to users and patients and validated through rigorous testing protocols rooted in scientific principles. These solutions take time, expertise and multidisciplinary collaboration, the Simpson et al. article being an excellent example. Our front-line healthcare workers and patients deserve nothing less.