Study Design, Participants & Setting
This study was a parallel-arm, triple-blind, randomized controlled trial. Ethical approval for the trial was obtained from the Institutional Review Board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster (#UW 21-129) and the clinical trial was registered with the Chinese Clinical Trial Register (ChiCTR2200064784). Participants were recruited from Philip Prince Dental Hospital and University Health Service Clinics by way of telephone interview and a brief clinical screening was performed for suitability to participate at the Institute for Advanced Dentistry Clinical Research Centre, HKU Dentistry. The inclusion criteria were (i) be aged 18 or above; (ii) no known allergy to mouthwashes; (iii) not taking or have taken any anti-viral, anti-bacterial or anti-fungal agents-topically or systemically in the past month; (iv) no contraindications to use of mouthwashes such as thyroid disease, or undergoing radioactive iodine therapy (which may interact with povidone -iodine).
Interventions
The three intervention agents were the Government recommended mouthwashes for use in the mitigation of COVID-19 and respiratory infections [26]: (i) Povidone-iodine 1% w/v, Betadine® (PVP-I), with instruction to rinse with 10mL for 30s and then spit out; (ii) Hydrogen Peroxide 1.5% w/v, Peroxyl®-Colgate® (H2O2), with instruction to rinse with 10mL for 1 min and then spit out; (iii) Chlorhexidine Digluconate 0.2% w/v, Corsodyl® (CHX), with instruction to rinse with 10 mL for 1 min and then spit out. The mouthwashes were administered by the attending nurse to patients at the Clinical Research Centre approximately 15 min prior to the assigned dental prophylaxis treatment (scale & polish).
Samples Collection & Outcome Assessments
Prior to administration of the intervention mouthwashes, unstimulated saliva samples were collected [Pre-Pre-Procedural Saliva Mouthwash Sample]. Participants refrained from eating, drinking or smoking for at least 30 minutes prior to sample collection. Whole unstimulated saliva samples were collected using the drooling method [27]. Standardised instructions were given to the patients to lean slightly forward on the dental chair and allow for saliva to pool in their mouth and not to swallow. Then to allow the saliva to drool into a 15 mL sterile centrifuge tube (Corning, USA). The time set for saliva collection was approximately 5 minutes to obtain 1-2 mL of saliva. The saliva samples were sealed and labeled and temporarily stored on ice in preparation for downstream analyses. This procedure was replicated following the administration of the assigned mouthwash [Post-Pre-Procedural Saliva Mouthwash Sample]. The harvested samples were dissolved in a viral transport medium (Working Earle’s BSS, 5%BSA, 20% glucose, working vancomycin, working Amikacin, working Nystatin and 4.4% NaHCO3) to maintain the viability and integrity of samples [28] in 5 mL Axygen transport tubes (Corning, USA) and they were stored in -80 ℃ (Haier Biomedical, China).
Environmental samples were collected in two ways (i) air samples and (ii) working surface samples. Air Samples were collected using an MD8 Airport Portable Air Sampler with standardised Gelatine Membrane Filters (Sartorius Scientific Instruments, China). The air sampler for the monitored area was set with a flow of 50 L/min and a total volume sampled of 2 m3. The air sampler was positioned at approximately 1.5 m above the floor and 1m horizontal distance from the head rest of the dental chair where the dental scaling was carried out, Figure 1. At the beginning of each clinical session prior to admittance of patients’ air samples were obtained as control air samples. The membrane from the filters was sectioned into small pieces using a sterile forceps and immersed into 2 mL viral transport medium in 5 mL Axygen transport tube, sealed and then were stored in -80 ℃.
Working surface samples of the dental chair tray and instrument holder were obtained by swabbing surfaces with a disposable specimen collection flocked swabs (Medico, China) on completion of dental treatment, Figure 1. Control samples were obtained from the dental chair tray and instrument holder prior to any dental treatment. The swabs were immersed in 2 mL viral transport medium in 5 mL Axygen transport tube, sealed and then were stored in -80 ℃.
Outcomes
The primary outcome is the reduction in the prevalence of respiratory pathogens as identified from the NxTAG® Respiratory Pathogen Panel + SARS-CoV-2 test (Diasorin, Italy). Assessment of the presence of ‘any’ respiratory pathogens is based on qualitative simultaneous detection of nucleic acids from multiple respiratory viruses in saliva samples. The secondary outcomes include the reduction in the prevalence of respiratory pathogens determined by air samples and swabs samples. The NxTAG® respiratory panel detects 19 viral pathogens including SARS-CoV-2: Adenovirus, Coronavirus 229E, Coronavirus HKU1, Coronavirus NL63, Coronavirus OC43, Human Bocavirus, Human Metapneumovirus, Influenza A, Influenza A – subtype H1, Influenza A – subtype H3, Influenza B, Parainfluenza 1, Parainfluenza 2, Parainfluenza 3, Parainfluenza 4, Respiratory Syncytial Virus A, Respiratory Syncytial Virus B, Rhinovirus/Enterovirus, SARS‑CoV‑2 (ORF1ab and M gene).
For the extraction of total nucleic acid, 10μL of MS-2 internal control was added to monitor the extraction and amplification steps, together with 200 μL of either saliva or environmental sample and mixed using Tianlong’s viral RNA and DNA extraction kit (T016H, Tianlong Science & Technology, China) in accordance with the manufacturer’s instructions. 60 μL elution volume of total nucleic acid (TNA) for each sample was acquired. For further testing, 35 μL pure extracted TNA was spiked into pre-plated NxTAG® Respiratory Pathogen Panel wells containing lypholized reagents and re-suspended gently on ice. Subsequently, the RT-PCR amplification was performed using an Applied Biosystems (Thermo Fisher Scientific, USA), with the detailed cycling parameters as one reverse transcription step at 42°C for 20 min; one template denaturation step at 95°C for 2 min and 30 s; 15 first amplification cycles at 95°C for 20 s, 65°C for 60 s, and 72°C for 10 s; 24 nested amplification cycles at 95°C for 20 s, 58°C for 60 s, and 72°C for 10 s; and a hybridization step at 37°C for 45 min. Subsequently, the plate was transferred to the 37°C preheated MagPix heater plate of the MagPix instrument (Luminex Corporation, Austin, TX). SARS-CoV-2 TNA and MS-2 were used for each single batch as positive and negative control, respectively. The total turnaround time was around 4h. The results were analyzed via Luminex SYNCT™ software 1.1. (Luminex Molecular Diagnostics, Toronto, Canada), Figure 2.
Sample size
In this study, the primary outcome is the reduction in the prevalence of respiratory pathogens by administering pre-procedural mouth rinsing. A previous study conducted in the University of Hong Kong identified that the prevalence of ‘any’ Respiratory Pathogen from saliva, as detected by NxTAG® Respiratory Pathogen Panel IVD was 23.4% among those with non-fever/ non-respiratory symptoms attending [29]. In determining the sample size/ sample power for this clinical trial the assumption was that patients attending the dental clinical would have a similar prevalence of respiratory pathogen (i.e. non-fever/ non-respiratory symptom patients 23.4%) and that ‘willingness to participate’ rate would be ~85% (as identified in a pilot study). Then in order to detect a significant change in the prevalence of ‘any’ respiratory pathogen following mouth rinsing a sample size of 196 would have 80% statistical power, using McNemar test (paired categorical data). To allow for ~15% drop-out it seemed prudent to recruit 228 participants.
Randomization of Participants
Randomization and allocation sequence were computer generated by an officer of the Translation Research Laboratory of the faculty who was not involved in recruitment or assessment of study participants and based on random digit generation with participants in blocks of three. Subject allocation was concealed in sequentially numbered opaque sealed envelopes (SNOSE) by an administrative assistant not involved with the study [30]. The attending clinical trial nurse opened envelopes after registration of the patient at clinical trial centre and provided the intervention (pre-procedural mouth rinse) in a room adjoining the dental surgery. In the present study, the participants were blinded to what mouth rinse they received as the mouth rinse were provided in a similar dispensing cup of similar volume (10 mL). The operators in the dental clinic and the research assistant collecting samples were blinded to what pre-procedural mouth rinse the participants had received as they were not present when mouthwashes were administered.
Statistical Analyses
Firstly, the data output from NxTAG® Respiratory Pathogen Panel was processed by the Luminex SYNCT™ software 1.1. (Luminex Molecular Diagnostics, Toronto, Canada) for (i) pre-procedural saliva samples, ii) post-procedural saliva samples, iii) environment control air samples (air samples prior to dental treatment), iv) environment air samples (post-treatment), v) control environment swabbed work surface (pre- dental treatment) and vi) environment swabbed work surface following treatment. The data on detection of the 19 viruses (detected (1), not detected (0)) was entered for saliva samples (pre and post), air samples (pre (control) and post) and swabbed-surfaces (pre (control) and post) into SPSS 27.0 IMB® SPSS® Statistics, USA). In addition, demographic details of participants were entered: age (continuous variable), gender (nominal/ binary categorical) and educational attainment level (ordinal categorical).
The demographic profile of participations among the three intervention groups were compared (employing Chi-Square tests for categorical variables and One-Way-Analyses of Variance (ANOVA) for age (continuous variable). Descriptive statistics were produced of the prevalence of each of the 19 type of virus (number, %), prevalence of ‘any’ detected viruses (number, %), and the number of viruses (range, mean (standard deviation)). The prevalence of ‘any’ detected virus in pre-procedural saliva and post-procedural via was compared using the McNemar test for paired categorical data comparisons. Change in the prevalence (∆) of ‘any’ detected virus in the pre-procedural saliva to the post-procedural saliva of participants was determined, and variations in the change in the prevalence (∆) between the three intervention groups was compared using Chi-Square tests. Following on, the mean number of detected virus in the pre-procedural saliva was compared to mean number of detected virus in the post-procedural saliva using the Wilcoxon signed-rank test (non-parametric equivalence of paired-t test). Changes in the number (∆) of ‘any’ detected virus in the pre-procedural saliva to the post-procedural saliva of participants was determined, and variations in the change in the number (∆) between the three intervention groups was compared using Kruskal Wallis tests (non-parametric equivalence to ANOVA).
All environmental post-treatment swabbed surfaces were analysed for presence of (prevalence) of each of the 19 types of viruses (number, %), prevalence of ‘any’ detected viruses (number, %), and the number of viruses (range, mean (standard deviation). Comparisons to pre-treatment swabbed surfaces to be made if applicable (virus detected in post-treatment swabbed samples) using statistics as outline above. A random sample of environmental air samples (post-treatment) were analyzed for presence (prevalence) of each of the 19 types of viruses (number, %), prevalence of ‘any’ detected viruses (number, %), and the number of viruses (range, mean (standard deviation)). Likewise, comparisons to control pre-treatment environment air samples was made if applicable (virus detected in post-treatment environment samples) using statistics as outline above.