How COVID-19 testing could improve research into oral care interventions
Professor Martin Addy and Professor Robert Newcombe discuss how research into the impact of oral care interventions could be incorporated into COVID-19 testing.
COVID-19 is a respiratory infection caused by a member of the coronavirus family of viruses, SARS-CoV-2. Infections rapidly reached pandemic proportions, with high levels of morbidity and significant mortality in all populations, including the UK.
Early in the pandemic risk factors for infection became clear including: older age; underlying medical conditions; obesity; ethnicity; even gender. Clear explanations for predisposition to infection have not always been forthcoming.
What also became apparent was that large numbers of individuals could become infected with the virus but exhibit little or no symptoms. This is a clear problem regarding the spread of the disease. The consensus scientific opinion is that the virus can be spread from person to person directly or indirectly.
Direct spread is by droplets expelled from the mouth and or nose during coughing, sneezing, singing or even just talking. Indirect spread is from surfaces that may become contaminated by the virus droplets, which are then picked up by an uninfected individual making contact with the surface.
These intermediary surfaces, often called fomites, which can spread infections, has been known for decades, and include the hands.
As with many viral respiratory illnesses, notably influenza, the risk to the infected individual is the development of pneumonia. This particularly applies to the elderly. The infectious agent, however, is not by any means always the virus, but can be a bacterium derived from the mouth.
Capacity to limit COVID-19 spread
Although this scenario has been appreciated for many years, the dental team was reminded, in a letter to the BDJ (Sampson et al, BDJ, 228, 12, 971, 2020), of the importance of oral care to prevent superinfection of the respiratory system – usually by oral bacteria – in those suffering with COVID.
The risk of bacterial superinfection in individuals already ill with COVID-19 is clearly increased by the probability that oral care is adversely impacted. Bacterial numbers in saliva expand many fold.
A number of groups and individuals based or associated with academic institutions, universities and dental schools in the UK have reported, through different media outlets, that oral care products, notably, toothpaste and mouth rinses, have the potential to help limit the spread of COVID-19 through their potential anti-viral action.
A potential additional benefit clearly would be the prevention of superinfection through antibacterial activity.
The aim of this article is to consider the overall prevention strategies recommended by government and medical advisers. It will also consider how oral care products may provide additional benefits through anti-viral and anti-bacterial actions.
Available evidence for these actions will be summarised and possible study protocols for future clinical studies presented. These study designs aim to exploit the present testing methods used to identify coronavirus in the mouth.
Recommendations for prevention
Based on knowledge specific to COVID-19 and years of experience of microbial infections and their spread, experts throughout the world have made recommendations, directly or through government, on measures to slow or stop the spread of COVID-19.
These can be considered under two headings, although there is overlap. The first relate to healthcare professionals knowingly dealing with COVID-19 patients. It also concerns the use of personal protective equipment (PPE) to prevent infection of the carer.
Here is not the place to exhaustively discuss the subject. Except to note that the principle of PPE expanded into many professions, ranging from all medical and dental teams to care home workers and hairdressers.
Public health measures
Additionally, the principle evolved to not only reduce the risk of infection of the professional but prevent the transmission of the virus to others by the professional. The second relate to prevent the spread of the virus in the general community and are essentially public health measures.
These are: regular hand washing with soap, detergents or antiseptic solutions or gels; wearing a mask – particularly in confined spaces eg on public transport, in shops – and social distancing. Essentially, all three measures are directed towards, directly or indirectly, reducing droplet spread of the virus.
How effective these three recommendations are alone or combined in reducing viral spread is difficult to say with any degree of certainty. This is particularly the case as each lacks direct scientific proof in respect of COVID-19 spread. They rely on extrapolation from laboratory and surface decontamination studies.
This said, the recommendations are simple, safe, inexpensive and scientifically based, albeit incomplete. A further approach is mass testing for carriage of virus in the mouth and or nose, and isolation/quarantine for those testing positive. As this lies at the heart of this article, discussion will be left for later.
Potential of oral care products
As stated, several articles over several months have been written on the potential of oral care products in COVID-19 infection. In fairness to the media, through dental and newspaper publications, this potential for oral care products in the pandemic has been well received and publicised at home and around the world.
Anecdotally, the same can be said of the dental team acceptance of the concept. Unfortunately, the same cannot be said of the powers that be. Namely the politicians and their medical advisers: perhaps because the publications have never reached their desks. This is despite a carte blanche acceptance and recommendation of hand washing to help prevent the spread of COVID-19.
It would appear that the evidence for hand washing to remove microbes, including coronavirus, from the hands is accepted as proven. But that potentially the same effects of oral care products in the mouth, are not. This is despite the fact that the evidence for the antimicrobial action of agents on the hands and in the mouth are broadly the same.
At this juncture, it would seem pertinent to outline the step-by-step approach of research. This would help to build an evidence-based proof of principle of benefit to human diseases and conditions.
Here, the focus will be on research protocols for human infectious diseases including COVID-19.
Evidence-based approach to control of infections
As the discussions for the main are concerned, with established formulations and products the important issue of safety testing will not be discussed. In respect of human, bacterial, fungal and viral infections there are three main steps in evaluating antimicrobial agents:
1. Laboratory testing (in vitro culture and sensitivity)
Simply, the microorganism(s) of interest are grown in appropriate media and challenged by the test agent(s). Kill, rate of kill, proportional kill, agent concentration and dose and other factors can be determined. As a general rule, but not absolute, agents that do not destroy the test micro-organisms in vitro will not work in vivo. Conversely, an antimicrobial effective in vitro may not work in vivo.
2. Surface disinfection studies
In view of the possibility/probability of indirect spread of infections via fomites, alluded to earlier, surfaces can be contaminated with a variety of micro-organisms of interest and challenged by antimicrobials. The same variables as discussed can be fed into the test protocol.
3. Randomised controlled clinical trials (RCTs)
In therapeutics, and for decades, RCTs have been the backbone for proving efficacy of agents used to improve health or wellbeing. Study designs can be parallel or crossover. With a parallel group design subjects are randomly allocated to a test agent(s) and a similar number to a control group. The control agent is formulated to have no expected therapeutic action. A recent example of such a parallel design is that carried out to evaluate a COVID-19 vaccine: one group receiving, by injection, the vaccine and the control group a dummy or placebo solution.
In a crossover design all subjects receive the different test formulations and the control during different test periods. In both parallel and crossover studies, blindness where possible is used. Thus, the person providing the formulations and/or the examiner recording results should not know what the subject is taking (single blind). The subject does not know what they have received (double blind).
This very brief precis of research methodology should now be applied to knowledge of approaches to control microbes on the skin and in the mouth with particular reference to the spread of COVID-19.
Wash hands, wear a mask, socially distance, brush teeth and mouth rinse: the evidence
A plethora of data is available from studies in vitro over many years to the present day. They show that numerous micro-organisms, normal and safe to pathogenic, that colonise our skin – including the hands – are rapidly destroyed by a range of chemicals. This includes soaps, disinfectants, detergents and antiseptics.
These include action in vitro against coronavirus. Surface disinfection studies also have shown broad spectrum activity against the same micro-organisms including coronavirus and various other virus. These studies did reveal the importance of concentration of the agent on effect.
An example relevant to this paper was chlorhexidine. Whilst ineffective against coronavirus at 0.02%, it was very effective at 0.05% and above: more in line with concentrations in rinses and hand washes. Clinical studies for antimicrobial action of hand washes are also quite numerous, possibly driven by proving the importance of hand washing before surgical procedures.
For some hand scrubs, action against recolonisation by bacteria persisted for several hours. RCT antiviral studies on hands are limited and for coronavirus not reported. Clearly, such studies might be difficult to perform, but leaves acceptance of hand washing in COVID-19 on extrapolation of data. Given the present circumstances this would seem eminently acceptable. But how to overcome this deficiency will be discussed later.
Wear a mask
The value of masks in reducing the spread of infection has long been debated, even regarding their value in an operating theatre. This may be an explanation for the government’s delay in recommending them in confined environments.
It is difficult to conceive laboratory models or clinical trials to test the hypothesis of whether masks are of any value or not as a public health measure in COVID-19.
Simulations of droplet spread with and without a mask have suggested benefits. Given the discussions over many years, and with the lack of any suggestion of a downside to wearing a mask, unless exempt, biology suggests some benefit: best to comply.
This appears to have been the basis for lockdowns and quarantine of possibly infected or infected individuals. It is also the idea behind ‘Test and Trace’. Both the spikes and falls in infection rates appear associated with social distancing.
Epidemiologists have reported from early in the pandemic that ‘hotspots’ of infection arose after mass gatherings in the UK. Abroad, massive lockdowns involving whole cities and regions had huge effects on transmission. Therefore, it is difficult to deny the importance of this measure.
Toothbrush and mouth rinse
Studies in vitro and of surface decontamination show that ingredients in toothpaste, notably detergents such as sodium lauryl sulphate, kill a range of micro-organisms, including coronavirus.
This is not a surprise since toothpaste detergents are the same as found in hand washing formulations recommended by medical experts and governments. Similarly, a range of antimicrobials in mouth rinses exhibit the same effects and include chlorhexidine, povidone iodine, cetylpyridinium chloride (CPC), peroxides and others.
It is worth pointing out that although many rinses contain alcohol, none, including essential oil rinses, have levels remotely high enough to kill bacteria and viruses, including coronavirus.
Perhaps, what has not been discussed in detail in discussions of oral care products in COVID-19 are the plethora of data for the antimicrobial action of toothpastes and mouth rinses in the mouth drawn from RCTs. As with hand washing, most studies are concerned with bacteria and can be described under two headings: 1. Salivary bacterial counts 2. Plaque regrowth.
1. Salivary bacterial counts
The mouth has at least 900 different species of bacteria and on average one millilitre of saliva contains 15 million bacteria. In these studies a saliva sample is taken early morning after a pre-study, overnight suspension of tooth brushing or rinsing.
The test product(s)/agent(s) or control agent is rinsed and saliva samples taken at intervals throughout the day up to seven hours after the rinse challenge. For toothpaste, the challenge can be a slurry rinse or a brushing.
Bacterial counts in the saliva samples are then made. Many of these studies have been published. They reveal some mouth rinses produced large reductions in bacterial counts. This persisted for several hours: as an example chlorhexidine produced reductions of >95%, which persisted for the seven hour period.
Indeed, the first study showed effects lasted to 12 hours. For toothpaste, bacterial numbers fell immediately and significantly compared to control, and persisted for three to five hours.
2. Plaque regrowth studies
Bacterial plaque continuously forms on teeth and has disease potential for caries and gingivitis. Regular mechanical removal by regular tooth brushing is the mainstay of plaque control. But toothpaste and rinses can add adjunctive chemical inhibitory action.
This effect has been shown many times in plaque regrowth growth studies. Thus, from a zero plaque score, toothpaste slurries and mouthwashes used twice a day, without mechanical cleaning, compared to a control significantly reduced plaque regrowth, usually over a four-day study period.
This data clearly shows that many oral care products have antimicrobial action in the mouth, sometimes profound. As with hand washing, supportive in vitro and surface disinfection studies strongly suggest the same products, including toothpaste, should impact coronavirus in the mouth.
The missing data, for both hand washing and oral care products, are effects in vivo. Although case reports reveal large reductions in coronavirus in the mouth after a single rinse with antiseptic mouthwash products. The final, section will suggest research protocols, which could provide the missing evidence through RCTs.
COVID-19 testing and oral care interventions
In the UK and many countries, mass testing for coronavirus in the mouth and nose has been adopted. First used for individuals with symptoms of COVID-19, pre-operation hospital patients and front line medical staff. In the UK, this has now progressed to testing whole cities.
The basic test is based on a single swab of the mouth and nose. The test is binary; positive or negative, with the result obtained in 24 hours. There can be an ‘unclear’, when a repeat test is requested. This suggests there has to be a level of virus for detection.
What is disappointing is the testing protocol, particularly as laboratory facilities for the test have expanded to swabs being taken separately from the nose and mouth and hand swab tests. This would provide a whole raft of data relevant to COVID-19 and its spread. It would form a basic platform for the research protocols to be outlined here.
There was a recent piloting of a test providing a rapid result within minutes, based only on a saliva sample. This further opens up the possibilities for the testing of oral care and hand washing products in the fight against the spread of COVID-19: accepting that the accuracy of the quick test has yet to be established.
The research protocol for RCTs on the effect of oral care products on COVID-positive individuals is as follows. Note the very same protocol could be used for those testing positive for the virus on hands to evaluate hand washing products.
Test the effects
- All those testing positive for coronavirus would be invited to participate in the study. In those subjects shown positive by the rapid test, participation could be immediate. In those shown positive by the swab test, participation would be by invitation, possibly at home, 24 or more hours later. For the latter a second pre-intervention swab test would be performed
- Those agreeing to participate would be randomly allocated to use either the oral care product under test or a control/placebo. Examples could be a test mouth rinse versus a saline rinse. Alternatively a tooth brushed toothpaste versus tooth brushing with saline etc
- A second swab would be taken shortly after the use of the intervention. In the case of the rapid test, the oral care product protocol would be applied immediately at the test centre.
This basic protocol could be used to test a range of oral care products to determine their immediate effects on the oral carriage of coronavirus. If applied to the hands, the same principle of hand washing effects could be tested. If the data indeed showed significant effects of oral care products on virus carriage, the time frame of testing could be extended to assess the duration of effect.
As with all RCTs, statistical advice should be sought. Not the least of which is the determination of sample (subject group) size, particularly given the binary nature of the outcome measure. Additionally, inclusion/exclusion criteria for subjects need to be decided. Balancing of test and control groups should be considered, if not only for age and gender.
Available evidence suggest that some mouth rinses and toothpaste have antiviral actions that might impact the spread of COVID-19. Antibacterial action should reduce the risk of superinfection in those with COVID-19. The suggested protocols for RCTs should provide the missing data for efficacy in vivo.
Martin Addy, emeritus professor, dentistry, University of Bristol.
Robert Newcombe, emeritus professor of biostatistics, Cardiff University.