Rapid clearing of aerosol in an intubation box by vacuum filtration.

Editor—Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a highly infectious respiratory pathogen disseminated by droplets and aerosols. Healthcare providers (HCPs) performing aerosol-generating procedures (AGPs) on coronavirus disease 2019 (COVID-19) patients are at risk of infection. AGPs include, but are not limited to tracheal intubation, extubation, mask ventilation, tracheostomy, oropharyngeal/tracheal aspiration, high-flow air/oxygen delivery, bronchoscopy, esophagogastroduodenoscopy, transoesophageal echocardiography, defibrillation, chest compression, and a range of dental, head and neck, and thoracic surgeries. Variations of Lai's aerosol barrier for limiting healthcare provider exposures has been rapidly adopted, but remains incompletely validated. , Cubillos and colleagues reported qualitative results of vacuum filtration, but clinically actionable time-to-clearance information is lacking. Efficacy of particle elimination by vacuum relates to air flow rates, which can be diminished by in-line viral filters essential to decontamination of outflow. Therefore, empirical testing is needed for each vacuum/filter configuration attached to intubation boxes to determine the particle elimination kinetics. Here, we present experimental data on the time course of active aerosol removal, comparing our hospital in-wall suction system and two low-cost commercially-available vacuums using an intubation box.

Our two-piece design intubation box (Fig. 1 a) includes active aerosol removal by attaching a vacuum with an in-line high-efficiency viral filter (Draeger SafeStar55R, German company). Aerosol removal by such filters could mitigate virus dispersion; this filter has 99.9999% viral filtration efficiency. We tested two vacuums, with stated air flow ratings of 60 cubic feet min−1 (CFM; Shop-Vac #9303511) or 23 CFM (Intex (Long Beach, CA, USA) mattress inflator/deflator #66639E), attached via standard airway circuit tubing. Separately, we also attached our hospital wall vacuum through a pressure regulator (Ohio Medical PISA, Gurnee, IL, USA) set to maximum (0.13 kPa) to a 2 L suction canister, then to the filter and box. In our practice, the patient is covered with a sheet or surgical drape (Fig. 1b).

Fig 1

Intubation box with improved mobility and vacuum filtration. (a) The two-piece intubation box with a vacuum and in-line particulate filter is shown as a schematic with overall dimensions shown, with red arrows showing detachable top. (b) A mock intubation setup is shown with the working window sealed with a gown (disposable) clipped into place, affording proceduralist arm mobility, aerosol enclosure, and vacuum elimination. The gown can be easily detached during airway rescue. (c) Aerosol elimination follows exponential decay kinetics, with hospital wall vacuum and two commercial vacuums improving clearance kinetics. (d) Vacuum aerosol removal significantly decreases particle clearance half-lives from 3.4 min (passive) to 1.0 min (wall suction), and to 0.28 min with the 23 cubic feet min−1 (CFM) vacuum, or 0.14 min with a 60 CFM vacuum. Time series from replicate experiments from 1c were fit to exponential decays after normalisation, and average half-lives (t1/2) were analysed by one-way analysis of variance (anova) (F(3,9)=52, overall P<0.0001). Aerosol clearance was significantly hastened with suction from the wall vacuum, and with the 23 or 60 CFM stand-alone vacuums vs passive clearance. ∗∗∗P=0.0001, ∗∗∗∗P<0.0001, anova with Tukey's multiple comparisons testing, error bars represent standard deviation.

To simulate viral aerosol contamination and clearance, an aerosol particle generator (TSI 8026, Shoreview, MN, USA) was placed inside the covered 35×45×50 cm plexiglass box. An aerosol particle counter (TSI PortaCount 8048) was connected to a 135 cm long sampling tubing inside the box. To measure baseline particle clearance without vacuum applied, we created a stabilised elevated particle count (2.5–6×104 particles cm−3); the particle generator was then turned off and particle count data sampled at 15 s intervals in technical replicates. For active aerosol removal, the vacuum source was turned on at the moment when the particle generator was turned off. Normalised counts were fit as exponential decays (r >0.95, Matlab, Natick, MA, USA) and half-lives analysed by one-way analysis of variance (anova) (Prism 7, GraphPad Software, San Diego, CA, USA) with significance set to P<0.05 and Tukey's post hoc pairwise comparisons test.

The 3.4 min half-life baseline aerosol clearance was reduced to 1.0 min with the wall vacuum, 0.28 min with the 23 CFM vacuum, and 0.14 min with the 60 CFM vacuum (Fig. 1c, one-way anova, F(3,9)=52, overall P<0.0001). The two stand-alone vacuum configurations were not statistically distinguishable (P=0.97), though clearance half-lives for each vacuum were shorter than with no vacuum (Fig. 1d, ANOVA post hoc Tukey's test: P=0.001 for passive vs wall suction, P<0.0001 for passive vs 60 CFM, P<0.0001 for passive vs 23 CFM).

We applied a vacuum and viral filter to an enclosed intubation box and determined aerosol clearance times in order to establish parameters for time-to-removal after use. Enclosed boxes with vacuums capable of filtering SARS-CoV-2 dispersed during AGPs are likely safer compared with intubation boxes open to the room. The National Institute of Occupational Safety and Health (NIOSH) ‘hierarchy of controls’ prioritises engineering and administrative controls over personal protective equipment (PPE) for mitigating occupational hazards, and PPE is considered the least effective (albeit indispensable) control. Although we promote this engineering control, proper PPE is still recommended despite any additional benefits offered by our system.

The Occupational Safety and Health Administration (OSHA) recommends US operating rooms maintain a minimum of 15 air changes per hour, equivalent to 99% aerosol removal in 18 min. Both 23 and 60 CFM vacuum pumps reached 99% clearance of the box in 90 s, and likely reduce collateral contamination of other operating room equipment. The reusable 23 CFM vacuum costs $20, and could save several hundred dollars in operating room time per use. Our hospital wall suction significantly reduced clearance times also, but flow rates for wall suction are not routinely controllable nor determinable in clinical practice, precluding broad extrapolation. Aerosol levels outside the box were not assessed, but gases suctioned through a viral filter with 99.9999% efficiency exceed recommended air quality regulations. For longer procedures necessitating aerosol removal, ear plugs should be used and pressures considered. Improvements towards lightweight barriers, disposable barriers, or both combining various features can be readily envisioned. Our design may afford improvements in proceduralist mobility restrictions and emergency access to patients, though further testing is warranted to verify patient safety. Improvements in control of perioperative inhalational risk may be an unexpected lasting impact of the COVID-19 pandemic, in the same way that universal precautions emerged from the HIV epidemic.

Authors' contributions

Concept and design: TI, SH, GC

Acquisition, analysis, and interpretation of data: TI, SH

Drafting of the manuscript: TI, SH

Critical revision of the manuscript: all authors