NTSP Specials (Season 2): Prof Tim Cook discusses risks to healthcare staff during pandemics

Tim Cook is a consultant in Anaesthesia and Intensive Care at Royal United Hospital Bath and a Professor at the University of Bristol, UK. Tim is an international expert in airway management and when the COVID-19 pandemic struck, Tim wrote widely on the risks to staff and developed guideliens to keep staff and patients safe. Tim spoke at the 6th International Tracheostomy Symposium, held in Manchester UK, in October 2021. This presentation is an extract from that meeting.

The UK National Tracheostomy Safety Project (NTSP) is committed to providing education, information and resources to improve patient safety and the patient experience for those with tracheostomies and laryngectomies. All of our resources are housed on our website www.tracheostomy.org.uk, accessed by over 30,000 visitors each month from around the world.

This is the only podcast to bring you literature reviews, hot topic discussions and interviews with healthcare staff, patients and families.

Our goal is to improve the safety and quality of care for patients with tracheostomies and laryngectomies through education. We work closely with patients, families and healthcare professionals to develop new resources to improve care. We’ve collaborated with key stakeholders in tracheostomy care since 2009, and developed freely accessible resources, supported by online learning developed with the UK Department of Health. We’ve worked with the Global Tracheostomy Collaborative since 2012 to improve care for patients and their families everywhere.

We are funded by grants, donations and in partnership with medical device companies through unrestricted awards. This podcast series is supported by unrestricted education funding from the Atos Learning Institute. The funding supports the professional production of the podcasts and videos, and the medical device companies that support us do not have any creative influence over the content that we record. All of our work is undertaken by volunteer healthcare staff, patients and their families. 

Most of our content is supported by videos. You can access our training videos and resources for Basic Care, Emergency Care and Vocalisation & Swallowing. Download and print bedhead signs and emergency algorithms from our resources. 

You can support our work by watching or clicking any of the advertising links that appear via the NTSP YouTube Channel. You can also donate directly to the NTSP through the NTSP website, or by clicking the Buzzsprout podcast hosting "support" links. You can support our work by watching or clicking any of the advertising links that appear via the NTSP YouTube Channel.

Transcript

SPEAKER_00 0:08

This episode is part of a series that was recorded for the October 2021 International Tracheostomy Symposium. I'm going to hand it over to the presenting team at the symposium, and we're going to hear from Professor Tim Cook. Tim is an elistist who works in Bath in the United Kingdom, and at the past the pandemic, Tim did an awful lot of work understanding the aerosols generated when we manage airways and tracheostomies and putting that in context for risks to healthcare workers. So I'm going to hand over to the team who were there on the sofa at the ITS. Tim is a consultant in anesthesia and intensive care medicine at the Royal United Hospital Bath. He is an honorary professor of anesthesia at the University of Bristol. He's the Royal College of Anaestists Director for the National Audit Projects, and he's a college advisor for All Matters Airway. So I'm going to hand over to Tim Cook.

SPEAKER_01 1:12

I'm going to talk about aerosols, tracheostomy safety, healthcare workers, but everything I talk about regarding healthcare workers also applies to others who come near to tracheostomyes. I have no conflicts. I would like to acknowledge my colleagues who have worked together, both with colleagues in Sydney at distance and in Bristol and the UK. And I'm not particularly tied to either dropleteer theory or aerosol theory. These are important colleagues that I've worked with in Bristol, both senior clinicians and senior university scientists. And that's important for any aerosol research. Summary of my talk is that coughs and sneezes spread diseases. This comes from the Spanish flu information campaign from the US government. If we take ourselves back to spring 2020, there was much emphasis on contact transmission from fomites, that is, virus that was lying on surfaces, for which there's not a huge amount of evidence, and for droplet transmission, so transmission of large particles, small, larger than five microns that were able to be, that were not able to be breathed deep into the respiratory tract and that generally travel no more than a metre from the patient. And a particular emphasis that the only circumstances in which aerosols were relevant, so that is that spread could be further than a metre or two metres, would be when aerosol generating procedures were undertaken. So we have this dichotomy into the sort of protection one needs for all non-AGPs, so that's outside particular procedures, is very basic PPE, and only higher level PPE you use, which includes a full gown, a respirator mask, eyewear, extra cleaning, and extra time spent between procedures or patients for these so-called aerosol generating procedures. And in the UK and in CDC, USA and elsewhere, a whole range of different, particularly anesthetic and intensive care procedures mentioned as AGPs, particularly relevant to the audience today, tracheostomy care and suction, both included. The context we must remember is that we're now dealing with a variant of COVID, which is both more infectious, more likely to cause hospital admission, and more likely to cause asymptomatic infection than previously. So the virus is roughly twice as dangerous as it was nine months ago. Some basic principles about aerosols, we have these three modes of transmission. We see the red droplets relatively large, uh falling down under the effects of gravity, often within three meters, so the story goes, and then a line there, potentially becoming these fomites for contact transmission, and only this miasma or mist of aerosol, small aerosol particles, spreading further. The reality, of course, as we as we can see here with a sneeze, is that this travels seven to eight meters at 20 feet. That model simply isn't right. Worth remembering that as a sphere gets larger, its volume increases by its cube. So a particle of 100 nanometers in diameter can contain one virus, whereas one that's five times larger at 125, and ones that is considerably larger at half a centimetre contain a hundred billion times more. Um, and therefore could not contain a virus. Most of them are below one micron, which is a thousandth of a millimeter, and both of the most of the volume of what we exhale and most or cough out, uh most of the volume and most of the number of particles, as seen in this graph here, are well below this one micron cutoff. What's an AGP? Well, WHO defined an AGP as a procedure that generates an aerosol. They've also emphasised AGPs with an increased risk of pathogen transmission, which are really the ones we're important at. They emphasised when they made these definitions around the times of the SARS epidemic, a lack of evidence supporting them. But we are left with this linear model of procedure-specific risk leading to a risk of transmission, which then in turn leads to a risk of healthcare worker harm. And a lot of this current thinking is based on this meta-analysis and systematic review, or rather systematic review by TRAN and colleagues with WHO input, which used rather poor quality studies, that's all they had available to them. And they didn't look at all at aerosols, but what they looked at was the increased risk of infection if healthcare workers were involved in intubation, non-invasive ventilation, tracheostomy, and mask ventilation. And with those procedures, there was an increased risk of healthcare worker infection. But aerosols were not measured at any point. So, how can we model that this risk of an individual who's become infected with COVID or exposed to COVID, their risk of harm? And we have this model that the the individual risk is impacted by age, ethnicity, sex, and comorbidity. Well, which is important? Well, the answers are the three most important risk factors. They are age, age, and age. And this has been shown really very elegantly by David Spiegelholter, the excellent uh communicator and statistician. And if we see here, so this is how with age, the risk of death from COVID increases in a linear scale. But if we log that, we get a straight line, this is normal risk of death, this is risk from COVID. And for every year, there's a 12% rise in risk of mortality. So there's a huge impact of risk. So if we model three generations on that, so here's me, my son's 20. So my son, who is 20, has let's call it that risk, which is we'll call that one. And for me, a generation older, I have almost a tenfold greater risk. And then my dear dad, uh uh at the age of 90 um had a risk more than uh more than 100fold higher. So we've got roughly a tenfold or a thousand percent increase in risk for each generation of age. How do other factors compare? So we say five to six years, your risk increases, that's 200%. The equivalent of being male rather than female is to roughly increase your risk by about 70%. We'll call that four or five years. If you're non-white, again it increases your risk of four to five years, but comorbidity is much less. So the risk is age is much, much important more important than sex, then ethnicity, then comorbidity. For a healthcare worker safety, there's a lot of noise, there's a lot of fuss, there's a lot of anger around PPE. But we need to remember that the most important uh matters for uh reducing risk of disease transmission are elimination and substitution, vaccination, national lockdown, screening patients, closing hospitals to visitors. These are the most important, and PPE is the least effective but most focused on. So a few facts. Fact number one, risk of hospitalization compared to their community, healthcare. So this is in uh yeah, healthcare workers are two-fold more likely to be hospitalized, um, and that members of their own household are twofold more likely to be hospitalized. So, first we look at risk of infection, and this is the safer study from London in the peak of the first uh surge in April last year. They looked at 200 healthcare workers and they followed their PCRs and their serology, so that's their antibodies, over a period of two months. And 50, 20 point 20% of the individuals were um infected during the time era testing, and 45% became showed evidence of infection at some point during the study. So that's compared to about 15% of the population or 7% of the population, so threefold greater risk of infection than those in the community. And there are various studies around the country which essentially have shown the same results. So if we look at those, the multiples of healthcare workers versus community rate infection, it's between two and four, and mostly towards the three or four increased risk. Second fact, healthcare workers who die are disproportionately frontline and non-white. And there's a number of ways that this has been looked at. And I looked at colleagues, this first became apparent when newspapers were reporting uh multiple individuals who died from healthcare workers who died from COVID, and the faces were predominantly uh brown and black, and that is the case in many countries. We looked at that in a coordinated way and wrote it up in in one of the uh healthcare journals. So we can see uh the different proportions of ages in different specialties within the healthcare practice, and by working out the different proportions of who works in in healthcare in different specialties, we can work out relative risks. If we apply this to ethnicity, so non-white ethnicity accounts for 14% of the population in the UK, 21% of NHS workers. If we then look at the proportion of the population amongst nurses, uh support workers, doctors, uh, and others, the proportion that are non-white and the proportion that are that died of COVID, the proportion that died of COVID is dramatically higher in the non-white ethnicity, and particularly in first generation migrants. I know this is true in the US as well. If we look at the doctors in particular, what sort of doctors were dying in that first uh surge of COVID, they were predominantly surgeons, physicians, and GPs, people at the front line. Uh, you'll note there are no anaestysts and no intensivists, and the vast majority were male and of uh non-white ethnicity. Fact number three more effective personal practice equipment can protect and reduce healthcare worker cross-infection, and that also applies to patient cross-infection. So, this is Dr. Y Wen Lang, uh doctor in uh Wuhan, who told the world about uh COVID and died of it from a healthcare worker acquired infection. This is an early report from I think uh January or February from Wuhan last year, um, in which 41% of infections uh were thought to be hospital acquired, and three-quarters of those were amongst healthcare workers. In late January, early February, the the Chinese authorities mandated much more intensive PPE. And in this large uh report, they showed a dramatic reduction both in the frequency of healthcare worker infections, dropping down to almost tenfold, down to four percent, and also in the severity of those infections when they occurred. In this uh series, the Chinese and Wuhan looked at 400 healthcare workers who were there for several weeks of relief work. They were all involved in frontline care, and they all did at least one AGP. They were very carefully monitored and they looked at both their PCR and serology rates, and they found that compared to patients who mounted an antibody response, none of these healthcare workers mounted an antibody response. So there was no infections amongst these highly protected healthcare staff. Two papers unreviewed from the UK and both written by advocates of widespread and enhanced PPE with FFP3 stroke N95 masks, as am I. This first one in Cambridge. So during the latter part of 2020, there was an overall rise in community rates of infection and also in healthcare worker rates of infection. But towards the end of the year, they changed their policy so that on the infected red wards, um, FFP3 masks were used. And what we see is a reduction in these epidemic outbreaks of healthcare worker infection on those red wards, and it becomes more akin to what's happening on the non-infectious wards. And they reported a four more than 40-fold decrease in the rate of um of healthcare worker infection on COVID wards as a result of switching practice. And then this recent study again by Tom Northman colleagues, and what they've looked at is the the impact of staff wearing FFP3 masks on hospital um uh acquisition of COVID amongst patients. So in the UK at the peak of the second surge, 20% of all patients in hospital with COVID had acquired it there, and it's now about 5% and rising towards the end of this year. But if we stretch if we separate um hospitals out into those that only used FFP3 masks for AGPs during alpha and delta, and compare those to which to use FFP3 masks more widely, the rate of hospital-acquired infection in patients, never mind in staff, is reduced. So presumably that infection is spreading through staff and asymptomatic, and this is an important finding. So we have this from a number of studies. This is a large study looking at 10,000 healthcare workers in Oxford, and they identified all sorts of risk factors for uh increased infection amongst healthcare workers, but those working in ITU had half the risk of everywhere else in the population. And we can see that anesthesia and intensive care, very low risk in sort of this heat map compared to uh to other specialties. So something about what we're doing in ITU, which is protective. This is borne out in a study from uh Birmingham when the the risk of being exposed of getting uh COVID in ITU staff was half that if you're a housekeeper or worked in uh acute medicine. And in terms of hospitalisation, going back to that earlier fact, this study from Shah, which looked at all healthcare workers in Scotland, showed that compared to non-patient-facing healthcare workers, patient-facing healthcare workers had a threefold increase in risk of hospitalization with COVID, and that extended to their household members who had an almost twofold increase in risk. Again, anaesthetist or rather intensive care staff, lower risk, half the risk, compared to those doing AGPs outside ITU. And we've looked at this in some detail in terms of the impact of mortality on intensivists and anaesthetists. And this is a rather complicated table, but essentially what we looked at was the number of deaths within each group, be it doctors, patient-facing doctors, anaesthysis intensivists, NHS workforce, or the whole population, and compared how many anaestyss there were compared to how many there'd be expected, and comparing the observed to the expected mortality amongst anaesthetist intensivists, whichever way you slice it up, anaesthysis are underrepresented either two or fourfold. So anaesthysis intensivists, low risk. Um, we also looked at um the specialties mortality by specialties in a different way, looking at different um uh publicly available data and showed that those um those those um occupations, healthcare occupations most highly associated uh with mortality from COVID were patient transport and paramedics. Those people going into patients' homes and transporting them early in their illness when they're most infectious. And then nurses and midwives on the front line and support staff and rather less medics, and as you become more senior, you become lower risk. And if you're not patient-facing, in fact, your your risk is overall less than the general population. So to summarize, healthcare workers staff are at significantly increased risk of infection, hospitalization, possibly death, and anesthesis intensive at lower risk compared to other healthcare workers. Next fact most designated AGPs simply are not aerosol-generating procedures, they don't generate aerosol. And there are a number of ways of studying this. You have to be in an ultra clean either theatre or a box or chamber that makes you uh clean so you can detect the noise over the sorry the signal over the noise of the local environment, and you have to have a high fidelity detector which both uh uh detects uh aerosols and puts them into bins by size and time of collection. And a series of papers from North America, from Australia, and from the UK have, in my view, adequately demonstrated that all these AGPs are not AGPs, and of all the ones studied, the only one that is an AGP is upper GI endoscopy with coughing or burping. It's essential to emphasise that coughing is the prime aerosol generating, along with sneezing and exertional respiratory activity. This has been shown in at least numerous studies, but these two that I was critically involved with, this is Nick Wilson's um from Sydney, and compared to breathing, which it detects more aerosol than tracheal intubation or than inserting a face mask uh is inserting a synchronotic airway. Coughing uh increased the amount of aerosol detected 400fold. And in uh Jules Brown's and uh Andy Shrimpton's study in Bristol, this is a cough, the amount of aerosol produced over uh in over 12 seconds in a cough, and intubation you simply can't detect it. If you look at numbers, there's effectively a uh a three orders of magnitude increase in in aerosol in cough compared to intubation. So coughing par exemplar the aerosol generator, and we need to think about people, about individuals and their risk, not about procedures. Briefly to mention tracheostomy, there are just a small number of studies that have looked at tracheostomy. This is one from the aerator group in Bristol, and the different phases of aerosol detection are here, and essentially there's there's no aerosol uh shown at all. This is an index cough uh except for when cleaning and draping. So nothing going on with the airway. The airway is completely closed and intact, and actually, this is artifact, this is opening swabs or opening packets or moving the patient, and that generates artifactual aerosol. So no aerosol during the actual phases of airway management, including all these in the airway phase and the tracheostomy phase. No particles. And a similar study uh or case report from Forbes McCain in in Melbourne, and really the only time that they identified any aerosol really was uh when the surgery was done and they did electrocautry. And there's a good report uh from JAMA which shows that electrocautery does not uh it actually kills the virus, and therefore there is no risk uh from aerosolisation during electrocautering. Let's mention that here. So here we have electrocautery which is undertaken on a raw chicken breast mixed in uh with blood and uh the actual SARS-CoV-2 virus. This was then collected on gel plates and attempted to grow. And what this study shows is that they were able to detect, they were able to capture the plume from the cautery, but nothing was able to grow. So there's no growth whatsoever using two different modalities and whatever mode of uh of cautery they used. Uh but if they just sprayed some blood into the air, uh they were able to dissect it. And note this is a log scale, so this these are each orders of magnitude difference. Finally, a rather disappointing paper from JAMA very recently. Uh, this uh report uh claims to look at aerosolisation during tracheostomy and tracheostomy care. But unfortunately, it's a mannequin and pig model study, it's not undertaken in a clean room. I don't think the methodology for detecting a signal over noise is adequate. Um, and I think the, for instance, to create an artificial cough, they simply squirted some air into a fluid-filled tracheostomy tube. I don't think that mimics uh human aerosol generation. So unfortunately, I think most of the results and most of the information in it is rather poor. The only thing I would take from it is that they they did show that covering the tracheostomy with almost any form of mask reduces the amount of aerosol detectable uh down to about a quarter or perhaps down to a tenth if you use a surgical mask or HME. So, whatever you cover the tracheostomy with, you as with any human airway, you dramatically reduce uh the amount of aerosols produced. They uh inaccurately stated that um during surgery on their pig model, uh, this created aerosolised particles, as we've seen from the previous studies, is irrelevant. So, what do I take as lessons for tracheostomy care? That any tracheostomy care that generates coughing or marked exertional breathing will spread aerosols to the close locality. Covering the tracheostomy with almost anything will reduce that. The more effective the barrier, the more effective the mitigation will be. And pretty much anything that doesn't lead to coughing or exertional breathing is highly unlikely to spread aerosols. My overall conclusion is more broadly: most AGPs don't generate aerosols. Cough and exertional activities absolutely do. Masks reduce spread, and mask respirators reduce spread and protect the wear as well. And so the persistence of these arguments that aerosol generated by medical procedures is higher risk than by cough is simply not based on science and is not sustainable. This model has to go. We have to have this more complicated model where we think about the infectivity of the patient, perhaps procedural event specific risk, our duration of contact and proximity to the patient, mitigation through these three factors, and then the overall susceptibility of the patient with age being the most important. And only when we've interact we've had all these interactions, vaccination being the greatest risk data, can we work about risk to healthcare worker and those nearby? To summarize once again, coughs and sneezes spread diseases.

SPEAKER_00 21:52

Thank you very much. Thanks to team and the team for that really interesting insight. And we've learned a lot of lessons that are relevant. Beyond the pandemic, thanks to Tim's work and those of his collaborators. As ever, the views and opinions we discuss on the podcast are our own and don't necessarily represent those from our various employers. You can follow us on our social media channels and find out more episodes via our podcast web pages, YouTube, or wherever you get your podcast from. Thanks for listening and look forward to seeing you next time.