Skip to content

By Roger Highfield on

COVID-19: Towards a Clean Air Revolution

The pandemic has alerted the world to the threat of airborne disease. A new study has shown the value of clean air, and also how filtration can curb antibiotic resistance in hospitals. Our Science Director Roger Highfield talks to Cambridge based intensive care consultant, Vilas Navapurkar about its findings.

During the pandemic, the adoption of face masks in the UK was delayed by an undue emphasis on the evidence from randomised control trials, which initially was ambiguous, rather than insights into the mechanism of how COVID-19 spread.

We now know mask wearing is among a number of public health measures that can cut the incidence of COVID-19, by around 10% in the case of face coverings.

Lives were lost as a result of the delay in their adoption, according to Trisha Greenhalgh of the University of Oxford

‘The COVID-19 pandemic has exposed major gaps in our understanding of the transmission of viruses through the air”, remarked Lydia Bourouiba of the Massachusetts Institute of Technology and colleagues in the journal Interface Focus.

‘These gaps slowed recognition of airborne transmission of the disease, contributed to muddled public health policies and impeded clear messaging on how best to slow transmission of COVID-19.’

Images of homemade face masks, sent in by members of the public as part of the Science Museum Group’s COVID-19 Collecting Project

We now know that these viruses can also be transmitted through aerosols, which are tiny. While a human hair is approximately 10 microns — 10 millionths of a metre across — a droplet, such as the spray that come out of your nose when you sneeze, can be up to 10 microns wide and aerosols are even smaller, less than one micron in width. Whereas droplets fall to the ground in under 30 seconds, aerosols, can float in the air for hours and travel long distances.

Now a new study in the journal Clinical Infectious Diseases has shown how hospitals can use air filtration to cut the levels of circulating pathogens, not just SARS-CoV-2 but other viruses, fungi and bacteria, the latter including “superbugs” that are resistant to antibiotics.

‘Florence Nightingale would have loved this study,’ commented Natasha McEnroe, Keeper of Medicine, adding the museum has treasured objects related to her in its collection.

‘All of her work on hospital design and ward layout was about placement of windows to encourage ventilation, the layout of buildings, even distance between beds.”

‘She didn’t know the science behind it but she knew clean air and ventilation equals better health.’

I talked to co-author Vilas Navapurkar, Consultant in Intensive Care Medicine and Anaesthesia, The John Farman Intensive Care Unit, Cambridge University Hospitals. His edited responses are in italics.


Even before wave one in March 2020 we knew from colleagues and friends in Lombardy, Italy, and Wuhan in China, where the pandemic seemed to originate, that COVID-19 was airborne.

We knew that – it’s just people weren’t listening when it came to the airborne threat. Now, working with colleagues in Addenbrookes hospital, we have proved it is real.

Other colleagues had randomly tested health care workers and volunteers who had no symptoms (asymptomatic) in the hospital, and found the presence of SARS-CoV-2, so we knew that there was asymptomatic spread of COVID-19.

Running into wave two of the pandemic in 2020, we could also see the effects of introducing masks.

Colleagues in Addenbrooke’s found that when you use FFP3 masks, which can filter our very fine dusts, fibres, fumes and mists, you decrease acquisition of SARS-CoV-2.

In the worst affected wards, you cut the asymptomatic cases by more than half. But we could go further.


I’ve been working for several years with my colleagues Prof Stephen Baker and Dr Andrew Conway Morris on the diagnosis – using precision diagnostic methods – and the biology of pneumonia.

That included tests of the genetic code of pathogens using PCR (polymerase chain reaction).

We have further developed and put into routine service a PCR microarray that tests for 52 pathogens that are known to cause pneumonia (called a TaqMan array card).

We validated this cutting-edge method in separate work and it is now the gold standard for the diagnosis of suspected pneumonia in adult patients on ventilators in Cambridge University Hospitals.

In addition, we have validated a separate high throughput PCR which we are using in our research – the Fluidigm Biomark HD system which has additional capabilities and flexibility.

I also thought to myself ‘this is an airborne problem so why don’t we see whether we can capture the air bug and remove it?’ I got wind of the fact that there were filtration units in the hospital from separate companies, which combined ultraviolet light and HEPA filtration.

However, these air filtration units were in rooms that were not being used very much because of lockdown and were also in places where there was good ventilation anyway.

I wanted to see if these machines worked.  There was no evidence whatsoever beyond controlled laboratories filled with smoke or incense or dust or carbon dioxide that they worked, or removed bugs that cause illness in a hospital setting.

I then had a coffee with one of my colleagues who had some anthrax samplers from the Centres for Disease Control in Atlanta on loan to him, which have an amazing capability to capture really tiny airborne particles  

These cyclonic samplers from the National Institute of Occupational Safety and Health (NIOSH) suck air cyclonically into a carefully designed chamber where the air whirs around in a cyclone, allowing larger particles (four to 10 micrometres, millionths of a metre) to be pushed by centrifugal force on the sides of a large Eppendorf (a tube).

It then accelerates it through a narrow tube and then deposits, once again through centrifugal force, one to four micrometre particles on another Eppendorf and finally, exits through a filter where we capture airborne particles that are one micrometre in size or less.

We placed these cyclonic samplers strategically to capture air near and far from the air filters on a drip stands in the surge ward and surge ICU, where they sampled the air for four hours (any longer and they heat up too much and damage the RNA and DNA of microbes).  

We took the tubes to Stephen Baker, who then extracted the non-human RNA and DNA and ran these samples through the Fluidigm microfluidic circuits (think of them as miniaturised laboratories).

Stephen Baker, Andy Conway Morris and I have been working together and validating them for ages. It’s really fancy. Very few places in the world have this technology, and even fewer made it work.

But we did and it’s also part of our evolving contribution to studying the antimicrobial resistance crisis.

COVID-19 Home Test Kit, 2020, part of the Science Museum Group’s COVID-19 Collecting Project


In February 2021, we chose two surge areas. These are areas that do not normally take patients with infectious diseases but, when hospitals have no capacity, we have to open them up and use them. 

One was a surge ICU (as intensive care unit for patients on ventilators). 

These ICU patients have a deep lung disease, where the virus has migrated into the lung to cause severe injury, that needs artificial ventilation to prevent death. 

The other was upstairs, a non-ICU surge COVID-19 ward, where the patients are not on ventilators and are coughing, talking and breathing for themselves. 

We placed the air filters in these two areas for three consecutive weeks. 

The air filtration units were switched off for week one, on for week two and then switched off again for week three as a “washout period”.

Remember this was a real-world environment evaluation on real surge wards both ICU and a ward – neither designed to care for the patients we had to put in them.

All the normal activity took place – nursing and medical procedures, proning of patients, cardiac arrests, patients moving in and out, and whatever else.

The data very vividly shows that there was no detectable airborne SARS-CoV-2 in the surge ward when the machines were in operation.

And when you switch it off, the airborne virus comes back. When it came to ICU, where patients are on ventilators, there isn’t much virus in the air. 


I’m an ICU consultant and was involved in the pandemic from the word go.  I’ve been through several pandemics. Nothing as bad as this.

Thanks to the kindness and goodwill of colleagues, I quickly put together a team of engineers, scientists and doctors.

I cannot explain to anybody outside of an Intensive Care Unit the ferocity, the brutality, and a physical demand of the pandemic and what these colleagues did. 

Instead of going home, they put the kit and clobber on, went back in with me.

They came in early to set up and went in. They took all the cuts on your face from the PPE, the dryness, the fear, the anxiety and everything that we were going through.

This was an experiment done with no resource, with no spare time or availability by heroic colleagues who were kind enough to listen to me and then went back into battle COVID-19 when they should have gone to bed.

Whilst we were looking for SARS-CoV-2, I said: ‘Let’s look at everything else’, and we saw a load of other stuff in the air on the ward, such as other viruses, fungi such as Aspergillus that cause disease.

So, though there was not much SARS-CoV-2 in intensive care, because of the ventilators, there were lots of other airborne pathogens, notably micro-meter or less viruses and bacteria.

We found that filtration cut the number of pathogens significantly. 

I’ve been a consultant in intensive care for 25 years and before that a junior and I’ve seen thousands of patients with hospital acquired pneumonia, but I’ve never seen one get it from licking a bed sheet or snorting a curtain.

It’s from breathing in these spaces, so it’s obvious to me to be honest, that they’re breathing it in. So, what we’ve shown is that we can remove these pathogens from the air.

‘Yellow Doctors 4’ painting by Susie Hamilton, part of the Science Museum Group’s COVID-19 Collecting Project


The obvious question to ask is does it decrease hospital acquired infection?

So, we are engaged in contributing our capability to a big study in Addenbrooke’s to ask that question, though it is not simple.

Even if we breathe the same air in the same room, my risk of catching COVID-19 or any other infection is different to yours, depending on if you are a smoker, are on steroids and so on. 

I’m unclear whether it will show the benefits of filtration clearly because there’s lots of complicating factors and reasons to get an infection but it’s important that we test this, as otherwise it will be hard to persuade already cash-strapped hospitals to adopt such solutions.


Stephen, Andy and I want to see a “clean air revolution.” 

We should have clean air standards or air hygiene regulations just like we do for food and water.

My concern goes beyond hospitals, we need clean air standards for schools and care homes because that’s where the disease burden is.

We’ve opened a light in a really dark door that’s never been opened before and found all these new questions to ask.

We need to we agree what the maximum microbial burden in the air in public/business buildings should be, how we meet these standards and most importantly how we monitor these new standards.


That is a really important question that society and governments have to discuss.

When you go to a restaurant or a pub and you drink a glass of water, you expect the glass to have been cleaned and the water tested to ensure it’s clean. We need the same expectation for clean air.

There is more work to be done, however.

We need to do more to measure and understand the problem and I’m getting numerous requests to measure air from around the world.

We need to work out how much air to recycle, to remove and clean, and see how well this works in the real world.

I think it’s unrealistic to think any government is going to spend the mega billions required to re-engineer existing hospitals or build new hospitals with eight to 10 changes of air per hour per room.  

Ultimately, however, we have to be able to have air standards the way we do for food and water and monitor it just like we do for food and water.


We’ve got the gold standard now for diagnosing pneumonia and in Addenbrookes ICU they are using our micro techniques so that precision diagnostics lead to better antibiotic stewardship.

We can also measure resistant pathogens in the air – with microfluidic technology we can measure anything – and measure the decreased risk of airborne transmission.

Antibiotic resistance is medicine’s equivalent of climate change.

Obviously COVID-19 has been an understandably huge distraction, but I don’t think people realize what trouble we are in on the antibiotic front, where resistant bacteria are rendering our drugs useless.


I would also like to measure the bugs and the antimicrobial resistance genes in the sewage of a building, as a background radar or safety net.

Then, when you get spikes in the RNA or DNA of resistance genes you can then track back to where it’s coming from and say to whoever is running that ward that they should sort out their stewardship of antibiotics.


The latest picture of how far the pandemic has spread can be seen on the Johns Hopkins Coronavirus Resource Center or Robert Koch-Institute.

You can check the number of UK COVID-19 lab-confirmed cases and deaths along with figures from the Office of National Statistics.

There is more information in my earlier blog posts (including some in German by focusTerra, ETH Zürich, with additional information on Switzerland), from the UK Research and Innovation, UKRI, the EUUS Centers for Disease ControlWHO, on this COVID-19 portal and Our World in Data.