Project Details
Description
Aerosolisation of SARS-CoV-2 during clinical procedures is a major concern. The safe resumption of essential NHS services is impaired by the need to mitigate the theoretical risk of cross infection from procedural aerosolisation. This includes extensive pre- operative planning, use of personal protective equipment (PPE) throughout, and delays in patient/staff movement before, during, and after the procedure. There is currently little to no evidence on aerosolisation risk for many procedures, hampering national guidance and greatly reducing NHS capacity.
The AERATOR study will address a critical gap in evidence by quantifying the concentration, size and temporal and spatial dynamics of aerosols produced during routine medical & surgical procedures in different environments. This will focus on five clinical specialties particularly impacted by procedural aerosolisation: dental, orthopaedic, respiratory, critical care and ophthalmology.
This work comprises three workstreams:
Workstream 1: Within the Bristol Aerosol Research Centre (BARC) we will rapidly (within 4 weeks) validate instruments to study aerosolisation in clinical settings.
Workstream 2: Instruments will be moved into clinical settings and, using multiple instruments and sampling techniques, will measure aerosolisation dynamics and size across time and space.
Workstream 3: By using novel equipment, only available within Bristol, to levitate virus within a CL3 laboratory, we will investigate the survival of SARS-CoV-2 in aerosol particles and determine its infectivity.
The information gathered in this study will allow us to inform hospital trusts, policy makers and Public Health England regarding the safe and maximally efficient NHS working across multiple specialties.
Our team led on the dental procedures aspect of this study.
The AERATOR study will address a critical gap in evidence by quantifying the concentration, size and temporal and spatial dynamics of aerosols produced during routine medical & surgical procedures in different environments. This will focus on five clinical specialties particularly impacted by procedural aerosolisation: dental, orthopaedic, respiratory, critical care and ophthalmology.
This work comprises three workstreams:
Workstream 1: Within the Bristol Aerosol Research Centre (BARC) we will rapidly (within 4 weeks) validate instruments to study aerosolisation in clinical settings.
Workstream 2: Instruments will be moved into clinical settings and, using multiple instruments and sampling techniques, will measure aerosolisation dynamics and size across time and space.
Workstream 3: By using novel equipment, only available within Bristol, to levitate virus within a CL3 laboratory, we will investigate the survival of SARS-CoV-2 in aerosol particles and determine its infectivity.
The information gathered in this study will allow us to inform hospital trusts, policy makers and Public Health England regarding the safe and maximally efficient NHS working across multiple specialties.
Our team led on the dental procedures aspect of this study.
Key findings
Dental Procedures findings:
Team Members
Tom Dudding, Sadiyah Sheikh, Florence Gregson, Jennifer Haworth, Simon Haworth, Barry G Main, Andrew J Shrimpton, Fergus W Hamilton, Anthony J Ireland, Nick A Maskell, Jonathan P Reid, Bryan R Bzdek, Mark Gormley
Settings and volunteers
Aerosol was sampled during three routine procedures: scale and polishing, dental extraction and orthodontic debonding. Within these procedures suspected AGPs included high speed drilling and use of the triple syringe (3 in 1). We also aimed to address concerns around slow speed drilling, surgical drilling, as well use of ultrasonic scalers. 41 adult patients were consecutively recruited from waiting lists requiring either periodontal, oral surgery or orthodontic treatment. Each patient was contacted via telephone, received an information leaflet via post, and provided written consent on the day of treatment. Our study was performed in a surgical side room of a UK dental teaching hospital, with a background concentration maintained as low as possible using a portable air filter (at 180 particles/litre- similar to modern surgical operating theatres), providing the highest level of aerosol detection sensitivity possible.
Methods
An Aerodynamic Particle Sizer (APS) (TSI Incorporated, model 3321, Shoreview, NM, USA; detection range: 0.5–20 μm diameter particles) was used to measure aerosol. A custom 3D-printed funnel (RAISE3D Pro2 Printer, 3DGBIRE, Chorley, UK) made from polylactide, with a maximum diameter of 150 mm, cone height of 90 mm and a 10 mm exit port, was attached to the APS inlet using conductive silicon sampling tubing (TSI, 3001788), approximately 0.90 m long and 4.80 mm in diameter.
Baseline patient measurements
Baseline readings were taken from each participant including tidal breathing at rest (60 s), counting out loud (60 s) and three voluntary coughs. The funnel was positioned at source (as close to the mouth as possible), with the patient seated upright [20]. Baseline characteristics of patients were reported using median and range for continuous data, alongside counts and percentages for categorical data, stratified by specialty type. To assess differences among specialties, age and sex distribution was compared using one-way ANOVA and Fisher’s exact test respectively.
Patient dental procedure aerosol measurement
We conducted an initial pilot study to investigate the optimum position and orientation for the 3D-printed funnel when sampling dental aerosol. This was determined to be 22 cm from soft tissue nasion to the top of the funnel, at approximately 45 degrees on the patient’s left side (11 o’clock position). For every case, a full mouth examination was carried out using a dental mirror, followed by local anaesthetic administration when indicated. Each patient received 3-in-1 syringe air drying (30 s), water (30 s) and then combined air and water (30 s) applied to their all their teeth. When necessary, up to 3-minute intervals between procedural steps were allowed for background reading levels to stabilise. The remainder of the treatment session was dictated by clinical need. In keeping with real-world practice, wide bore high volume aspiration at 300 L min-1 was used during all procedures except oral surgery, for which Medi-VacTM suction at 60 L min-1, with a Yankeur suction tip was used. To measure aerosol generated by the dental instruments alone, we conducted high fidelity control experiments in triplicate, in a phantom head unit. For phantom head control data, the aerosol number concentration and size distribution were extracted for further analysis.
Statistical analysis
The aerosol number concentration for each procedure and baseline measurement were compared by calculating particle number concentration detected above background for each patient (irrespective of particle size). As the length of procedure differed across patients, we sampled the mean particle number concentration across the sampling time for each patient and the per patient values were combined to give median and inter-quartile ranges of total aerosol number concentration for each procedure.
All procedures were carried out with matched procedures conducted in phantom head controls, generating thousands of data points for comparison. Phantom control data represents the non-salivary contaminated aerosol from the instrument source, which poses no risk to patients or operators. Where aerosol was detected, we assessed whether the aerosol size distribution from patient procedures was explained by the non-salivary contaminated instrument source. Using this data, we were able to identify dental instruments with very low chance of ‘risky’ aerosol production and those which may potentially increase the risk of viral transmission by an aerosol route. The aerosol size distributions from the phantom head control and patients were compared, with the assumption that if the distributions were the same, all aerosol detected from the patient during the procedure could be explained by the non-salivary contaminated instrument source (represented by the phantom control). To further assess if size distributions between the phantom head control and patient differed other than by chance, a two-sided unpaired t-test was used to compare the mode widths.
Results Summary
Aerosol generating procedures (AGPs) are defined as any procedure releasing airborne particles <5 μm in size from the respiratory tract. There remains uncertainty about which dental procedures constitute AGPs, which has led to emergency mitigation measures, including enhanced personal protective equipment such as FFP3 (or N95) masks, ensuring adequate room ventilation and allowing additional ‘fallow’ time between patients to enhance aerosol dispersion. This resulted in a substantial reduction of dental patients able to access timely and effective dental care, with the British Dental Association warning of an “oral health crisis”. To our knowledge we conducted the largest clinical dental aerosol study during the pandemic, which aimed to accurately measure instrument aerosol ‘at source’ (i.e., as close as possible to the oral cavity). The study analysed a full range of common dental procedures on patients, using sensitive instrumentation and a rigorous scientific approach.
In total, fifteen dental procedures were assessed during periodontal, orthodontic, and oral surgery treatments. Of these, examination with a dental probe, hand scaling, local anaesthetic delivery, routine extraction (with forceps and/or elevator), raising a soft tissue flap, orthodontic bracket removal, alginate impression taking, 3-in-1 water only, and suturing did not produce any aerosol. For the other six procedures where aerosol was detected, the percentage of total procedure time that aerosol was observed was 12.7% for ultrasonic scaling, 24.8% for 3- in-1 air only, 75.3% for 3-in-1 air + water, 40.1% for high-speed drilling, 49.9% for slow speed drilling and 55.6% for surgical drilling. While high and slow speed drilling produced aerosol from patient procedures, which appear to have different size distributions from phantom head controls that may pose a risk, ultrasonic scaling and triple (3-in-1) syringe use did not appear to generate additional aerosol above that of the instrument itself and therefore does not increase the risk to dental teams, relative to the risk from being in close proximity to the patient.
Further detail on our results are available in the paper ‘A clinical observational analysis of aerosol emissions from dental procedures’ which has been published in PloS One. The information gathered in this study was used to inform hospital trusts, policy makers and Public Health England regarding the safe and maximally efficient NHS working in dental settings. We were also invited by the New Zealand Dental Council to advise them on the closure of dental services given they were experiencing peak infection rates at a later time than the UK.
Publications
Dudding T, Sheikh S, Gregson F, Haworth J, Haworth S, Main BG, Shrimpton AJ, Hamilton FW, AERATOR group, Ireland AJ, Maskell NA, Reid JP, Bzdek BR, Gormley M. (2021) A clinical observational analysis of aerosol emissions from dental procedures. Plos One https://doi.org/10.1371/journal.pone.0265076.
Team Members
Tom Dudding, Sadiyah Sheikh, Florence Gregson, Jennifer Haworth, Simon Haworth, Barry G Main, Andrew J Shrimpton, Fergus W Hamilton, Anthony J Ireland, Nick A Maskell, Jonathan P Reid, Bryan R Bzdek, Mark Gormley
Settings and volunteers
Aerosol was sampled during three routine procedures: scale and polishing, dental extraction and orthodontic debonding. Within these procedures suspected AGPs included high speed drilling and use of the triple syringe (3 in 1). We also aimed to address concerns around slow speed drilling, surgical drilling, as well use of ultrasonic scalers. 41 adult patients were consecutively recruited from waiting lists requiring either periodontal, oral surgery or orthodontic treatment. Each patient was contacted via telephone, received an information leaflet via post, and provided written consent on the day of treatment. Our study was performed in a surgical side room of a UK dental teaching hospital, with a background concentration maintained as low as possible using a portable air filter (at 180 particles/litre- similar to modern surgical operating theatres), providing the highest level of aerosol detection sensitivity possible.
Methods
An Aerodynamic Particle Sizer (APS) (TSI Incorporated, model 3321, Shoreview, NM, USA; detection range: 0.5–20 μm diameter particles) was used to measure aerosol. A custom 3D-printed funnel (RAISE3D Pro2 Printer, 3DGBIRE, Chorley, UK) made from polylactide, with a maximum diameter of 150 mm, cone height of 90 mm and a 10 mm exit port, was attached to the APS inlet using conductive silicon sampling tubing (TSI, 3001788), approximately 0.90 m long and 4.80 mm in diameter.
Baseline patient measurements
Baseline readings were taken from each participant including tidal breathing at rest (60 s), counting out loud (60 s) and three voluntary coughs. The funnel was positioned at source (as close to the mouth as possible), with the patient seated upright [20]. Baseline characteristics of patients were reported using median and range for continuous data, alongside counts and percentages for categorical data, stratified by specialty type. To assess differences among specialties, age and sex distribution was compared using one-way ANOVA and Fisher’s exact test respectively.
Patient dental procedure aerosol measurement
We conducted an initial pilot study to investigate the optimum position and orientation for the 3D-printed funnel when sampling dental aerosol. This was determined to be 22 cm from soft tissue nasion to the top of the funnel, at approximately 45 degrees on the patient’s left side (11 o’clock position). For every case, a full mouth examination was carried out using a dental mirror, followed by local anaesthetic administration when indicated. Each patient received 3-in-1 syringe air drying (30 s), water (30 s) and then combined air and water (30 s) applied to their all their teeth. When necessary, up to 3-minute intervals between procedural steps were allowed for background reading levels to stabilise. The remainder of the treatment session was dictated by clinical need. In keeping with real-world practice, wide bore high volume aspiration at 300 L min-1 was used during all procedures except oral surgery, for which Medi-VacTM suction at 60 L min-1, with a Yankeur suction tip was used. To measure aerosol generated by the dental instruments alone, we conducted high fidelity control experiments in triplicate, in a phantom head unit. For phantom head control data, the aerosol number concentration and size distribution were extracted for further analysis.
Statistical analysis
The aerosol number concentration for each procedure and baseline measurement were compared by calculating particle number concentration detected above background for each patient (irrespective of particle size). As the length of procedure differed across patients, we sampled the mean particle number concentration across the sampling time for each patient and the per patient values were combined to give median and inter-quartile ranges of total aerosol number concentration for each procedure.
All procedures were carried out with matched procedures conducted in phantom head controls, generating thousands of data points for comparison. Phantom control data represents the non-salivary contaminated aerosol from the instrument source, which poses no risk to patients or operators. Where aerosol was detected, we assessed whether the aerosol size distribution from patient procedures was explained by the non-salivary contaminated instrument source. Using this data, we were able to identify dental instruments with very low chance of ‘risky’ aerosol production and those which may potentially increase the risk of viral transmission by an aerosol route. The aerosol size distributions from the phantom head control and patients were compared, with the assumption that if the distributions were the same, all aerosol detected from the patient during the procedure could be explained by the non-salivary contaminated instrument source (represented by the phantom control). To further assess if size distributions between the phantom head control and patient differed other than by chance, a two-sided unpaired t-test was used to compare the mode widths.
Results Summary
Aerosol generating procedures (AGPs) are defined as any procedure releasing airborne particles <5 μm in size from the respiratory tract. There remains uncertainty about which dental procedures constitute AGPs, which has led to emergency mitigation measures, including enhanced personal protective equipment such as FFP3 (or N95) masks, ensuring adequate room ventilation and allowing additional ‘fallow’ time between patients to enhance aerosol dispersion. This resulted in a substantial reduction of dental patients able to access timely and effective dental care, with the British Dental Association warning of an “oral health crisis”. To our knowledge we conducted the largest clinical dental aerosol study during the pandemic, which aimed to accurately measure instrument aerosol ‘at source’ (i.e., as close as possible to the oral cavity). The study analysed a full range of common dental procedures on patients, using sensitive instrumentation and a rigorous scientific approach.
In total, fifteen dental procedures were assessed during periodontal, orthodontic, and oral surgery treatments. Of these, examination with a dental probe, hand scaling, local anaesthetic delivery, routine extraction (with forceps and/or elevator), raising a soft tissue flap, orthodontic bracket removal, alginate impression taking, 3-in-1 water only, and suturing did not produce any aerosol. For the other six procedures where aerosol was detected, the percentage of total procedure time that aerosol was observed was 12.7% for ultrasonic scaling, 24.8% for 3- in-1 air only, 75.3% for 3-in-1 air + water, 40.1% for high-speed drilling, 49.9% for slow speed drilling and 55.6% for surgical drilling. While high and slow speed drilling produced aerosol from patient procedures, which appear to have different size distributions from phantom head controls that may pose a risk, ultrasonic scaling and triple (3-in-1) syringe use did not appear to generate additional aerosol above that of the instrument itself and therefore does not increase the risk to dental teams, relative to the risk from being in close proximity to the patient.
Further detail on our results are available in the paper ‘A clinical observational analysis of aerosol emissions from dental procedures’ which has been published in PloS One. The information gathered in this study was used to inform hospital trusts, policy makers and Public Health England regarding the safe and maximally efficient NHS working in dental settings. We were also invited by the New Zealand Dental Council to advise them on the closure of dental services given they were experiencing peak infection rates at a later time than the UK.
Publications
Dudding T, Sheikh S, Gregson F, Haworth J, Haworth S, Main BG, Shrimpton AJ, Hamilton FW, AERATOR group, Ireland AJ, Maskell NA, Reid JP, Bzdek BR, Gormley M. (2021) A clinical observational analysis of aerosol emissions from dental procedures. Plos One https://doi.org/10.1371/journal.pone.0265076.
| Status | Finished |
|---|---|
| Effective start/end date | 7/08/20 → 6/08/21 |
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.