Thread by @ukhadds

I’m excited to discuss our latest research on how ambient CO2 affects how long #SARSCoV2 remains infectious in air. We report that even subtle increases in CO2 affects both how long #COVIDisAirborne and transmission risk. Here’s a🧵going over the findings
nature.com/articles/s4146…

Somebroader context: this study is the third part of our SARS-CoV-2 (SARS-Wars(?)) trilogy.

In our first paper we reported that the decay dynamics of the original strain. We reported that the viral decay rate is faster than initially reported: 90% of the viral load is inactivated in 20 min. We proposed aerosol pH was playing a role in the decay

pnas.org/doi/full/10.10…

In our second study, we systematically went through all of the various mechanisms proposed for viral decay in aerosol. We found of all the mechanisms,aerosol alkalinity played the largest role.

royalsocietypublishing.org/doi/10.1098/rs…

To help you better understand the background of the current paper, you may find it useful to learn about what got us to this point. I wrote a thread about it here:

When exhaled, the aerosol will reach a very high pH (>10) in 10s of seconds due to the dissolved CO2 (HCO3) leaving the droplet. The equilibrium shifts right, causing the acid (H3O) in the droplet to lower, resulting in a pH rise.

Elevated CO2 concentrations in the atmosphere limit the degree the equilibrium goes right. Thus, excess CO2 will affect the maximum pH the aerosol reaches. The virus is sensitive to the pH in this region (>10); any subtle change in pH can have a large effect in aerostability.

Implications: if the high pH of the respiratory aerosol is driving the loss of viral infectivity, it means that any increase in the conc of CO2 must have an effect of viral aerostability. The question becomes, will a difference be observable at CO2 conc in the real world?

We used a levitation/sampling technology we developed, called CELEBS (Controlled Electrodynamic Levitation and Extraction of Bioaerosols onto a Substrate), to probe the survival of different SARS-CoV-2 variants (Del+Omi) in mimics of exhaled particles in different conditions.

When compared to the Delta variant, the Omicron variant is more stable in a highly alkaline solution (pH 11). This relationship further supports our hypothesis presented in our previous publications that the high pH of respiratory aerosol is driving the loss of viral infectivity.

Omicron (BA.2) was found to be more aerostablethan the Delta variant when the humidity was high. In our previous study, we found that as the virus evolved, it had lost aerostability. This is the first time we’ve recorded an increase.

Omicron was also found to be more aerostablethan the Delta variant across a broad range of humidities.

Perhaps this played a role in Omicron rising to prominence?

We levitated the Delta variant in air that had various levels of CO2 and measured the infectivity. Any increase in the concentration of CO2 was found to result in an increase the virus’s aerostability. At higher concentrations, the difference became more significant.

It has become popular to use CO2 monitors to estimate risk. This is based upon the idea that the CO2 informs about how well a space is ventilated. From our study, the results show that the levels of the CO2 will also inform about how long the virus remains infectious in the air.

A moderate increase in the concentration of CO2 dramatically affected the decay dynamics. Changes in CO2 concentrations resulted in a completely different decay profile, where a difference in aerostability was observed within as little as 30 seconds, and throughout.

This data shows that the concentration of CO2 has a dramatic effect on how the aerosolized viral load will accumulate. Over longer periods of time at high CO2, the rate in which the virus becomes inactivated gets slower (after 20 mins only a subtle change in viral infectivity)

In terms of total aerosolized viral load, this means that after 40 minutes, more than 10x the amount of virus remains infectious in the air as a result of an increase in CO2 concentration. This will have a massive effect on the risk of transmission in a poorly ventilated space.

Increased concentrations of CO2 effected virus aerostability across a broad range of humidity. Below a humidity of 90%, the increase in CO2 roughly doubled the amount of virus that remained infective. This suggests that CO2 is more important than RH on aerosolized viral spread.

Using a Wells-Riley model, the time until there is a 50% chance that at least one susceptible person will become infected in a classroom containing a single infected person was estimated. Values are not absolute; rather, the trend of how CO2 conc affects the odds of transmission.

This study mechanistically connects the predicted rise in ambient levels of CO2 with an increased likelihood of respiratory disease transmission. As the global CO2 concentration continues to increase, the length of time respiratory viruses remain infectious in the air as well.

The expected increase in the concentration of CO2 in the atmosphere may also play a role in the emergence of new viruses. High CO2 concentrations will allow aerosolized viruses to remain infectious in the air longer, making it more likely to transmit from an animal to a person.

Indoor concentration of CO2 is highly influenced by the outdoor concentration, where the indoor CO2 concentration is usually higher. Thus, the increase outdoor ambient CO2 may have a much more significant effect on the indoor stability/transmission of the virus.

It has become apparent from our studies on SARS-CoV-2 (and MHV) that the decay of rate of respiratory viruses do not follow an exponential decay, where they have a single half-life. Rather, the profile is much more complex, where they appear to have a triphasic decay structure.

There are 3 distinct phases that we’ve termed:

Lag Phase
Dynamic Phase
Slow Decay Phase

The environmental parameters that affect the loss of viral decay in each phase is different. Knowing these differences will help to estimate the odds of short and long-distance transmission

Lag Phase:

In the Lag Phase the conditions in the droplet are not yet toxic to the virus. The length of the Lag Phase is dependent on humidity. At low humidity, the particle will crystallizes, causing about half of the virus to be inactivated.

Lag Phase:

At high humidity, the length of time of the Lag Phase is dependent on the pH resilience of the virus and may go on for minutes.

Dynamic Phase:

In the Dynamic Phase, the conditions in the droplet have begun to become toxic to the virus. During this phase, a large fraction of the virus is inactivated over a short period. At high humidity the pH is climbing, causing the droplet to be more toxic to the virus

Slow Decay Phase:

In the Slow Decay phase, the conditions in the droplet are becoming less toxic to the virus, where the pH is slowly being reduced by the trace amount of acid in the air. The rate of loss in this region is highly dependent on the acid content of the air (CO2).

Collectively, the triphasic decay profile of respiratory viruses affords new insights into how conditions affect transmission. It provides a framework to discuss how environmental factors such as humidity and acids will affect transmission through their collective effect on decay

For example, the Lag Period is highly affected by humidity, while the Slow decay region is not. Meaning that humidity would be expected to affect short range transmission and not long-distance transmission.

Conversely, CO2 concentration has a small effect on the Lag Phase and a massive effect on the Slow Decay Phase. Thus, the concentration of CO2 would be expected to affect long distance transmission of the virus and have only a minor effect on short distance transmission.

The slow decay phase measured using the CELEBS is similar to the decay rate measured using other techniques, such as rotating drums.

The results of this study reinforce the importance of both #ventilation and masks in limiting the transmission of respiratory viruses.

This research furthers our understanding of the environmental conditions that affect the transmission of respiratory viruses. It also directly links how the conditions that drive #climatechange also increase transmission risk.

This research was funded by BBSRC, EPSRC, and MRC.

I am very proud of our teams years of hard work on this project (Mara, Henry and Robert).

If anyone has any questions, I would be more than happy to (try to) answer them.

Thanks!

* I should have said "may allow aerosolised viruses to infectious in the air longer" than "will". Work needs to be done on many other viruses to explore the degree to which this can be applied across the spectrum.

@ScientistAndre The more robust the virus is to the pH change, the longer the Lag Phase. The absence of any viral decay during this short time period may increase the likelihood of short distance transmission.

@moog77 Until now, the focus has been on the effects of temperature and humidity on viral decay. We're establihing here that the more important parameter to consider is CO2. Over time, the exact amount of CO2 to be considered "clean" will be nailed down, but it will take some time.

@J__Doh Also, like you say, surface studies will use very high doses meaning they are measuring orders of magnitude of decay. We are measuring between 1 and 2 orders of magnitude.

@DRTomlinsonEP Essentially, it's a boot straps situation where we need to collect as much decay data, for as many microbes as we can, to really explore this space. I like the idea of looking at SARS1 and 2 to see how they differ.

@IzabelMolloy Thus it becomes a trade off: use a lot of energy to slighlty reduce the indoor CO2. The sense I got is that currently they are not economically feasible. I suppose if a market is created, all that could change. It would be an ideal solution.

@lzmddngs What is new here is the understanding that the CO2 itself will affect how long the virus remains infectious in the air. This is built upon our previous work where we identified the importance of aerosol pH in driving viral decay in the air.

@lzmddngs I understand what you are saying, but what we are reporting here is new, and likely, very important. CO2 is both an indicator of viral load, but also actively affecting viral stability.

@DtaGuy When we cut the volume of the droplets in half, we saw effect. That is not to say, as the droplets get smaller the reduction in pH may become more important.

@DrDeaRoberts @JJason_DJ_FM_AM @IzabelMolloy That said, I appreciate the guys ambition to try something bold.

@Mike_Honey_ I think it's important for people to think about risk in regards to this parameter, as you are doing here. Over time, we will nail down exactly where the hard lines are. To figure that out, we need people to understand this dynamic is important, and actively look into it.

@moog77 @PrasadKasibhat1 @CathNoakes @Lucas_Bolivian @linseymarr @jljcolorado @kprather88 @j_g_allen @CorsIAQ Figure 2B is the result of >1,000 individual levitations. To do the decay profile over 40 minutes at a single condition takes weeks in a Cat-3 lab. We just measured the curves at a few conditions to show the effect. The decay profiles at other CO2 needs to be done at some point.

@moog77 @PrasadKasibhat1 @CathNoakes @Lucas_Bolivian @linseymarr @jljcolorado @kprather88 @j_g_allen @CorsIAQ 2) There isn't a difference at higher CO2s at 2 minutes. Time is a critical factor in this. Over time, the buffering effect of CO2 will play a larger and larger role in causing the virus to live longer:

@moog77 @PrasadKasibhat1 @CathNoakes @Lucas_Bolivian @linseymarr @jljcolorado @kprather88 @j_g_allen @CorsIAQ 3) I think if you look at the above 2 figures, this may answer this one? Essentially, if we locked the time at 10 minutes rather than 2 minutes, figure 2A would look very different (somthing like this):

@moog77 @PrasadKasibhat1 @CathNoakes @Lucas_Bolivian @linseymarr @jljcolorado @kprather88 @j_g_allen @CorsIAQ We have enough information based on the data we've collected and presented to know that this is general trend that would occur. We also know why this trend would occur (relationship between slow pH change and viral pH sensitivity).

@moog77 @PrasadKasibhat1 @CathNoakes @Lucas_Bolivian @linseymarr @jljcolorado @kprather88 @j_g_allen @CorsIAQ Another issue is that the pH of a respiratory aeorsol is never static (over the ime periods relevant to disease transmission, ~an hour or two). It's this increasing followed by a slower decrease is going to make modeling this accurately a bit more challenging than a half-life

@moog77 @PrasadKasibhat1 @CathNoakes @Lucas_Bolivian @linseymarr @jljcolorado @kprather88 @j_g_allen @CorsIAQ For SARS-CoV-2, it seems like pH is about it for decay loss. For other viruses, there may be other factors that play a role, such as salt concentration, etc.

@Lucas_Bolivian @moog77 No doubt that diffusion will play a massive role. The issue is that when you have a large fraction of the exhaled dose not decaying, the total load in the room will elevated such that no amount of diffusion will matter; the virus will be everywhere.

@Lucas_Bolivian @moog77 To truly know one way or the other, I think the complex dynamics that we are reporting need to be coupled with comprehensive modelling, which is then compared against case studies to see what matters, what doesn't and to what degree.

@moog77 @Lucas_Bolivian As you say, there's many different mechanisms that reduce the load. I think the issue is where and when each one out competes the others.

@PrasadKasibhat1 @moog77 @CathNoakes @Lucas_Bolivian @linseymarr @jljcolorado @kprather88 @j_g_allen @CorsIAQ Here's the corrected curves:

@cain_rob You can see it here: ncbi.nlm.nih.gov/pmc/articles/P…

@akm5376 For example, NH3 could help lock the pH very high, killing the virus faster. Where as acids may have the opposite effect... To a point... If enough acid is acid, the pH of the aerosol may become acidic, which in turn would inactivate the virus.

@RowanKaiser 800 ppm was chosen as it is the level where air quality is deemed as "good".
Omicron can handle the pH change better than Delta. As such, under the same conditions, it will always do as good or better than the Delta. Note, the more pH sensitive the virus the larger the CO2 effect

@RowanKaiser Not sure if that answered your question, or just raised more...

@1goodtern To that end, the ability to quantify the cumulative acidic vapor load (the balance of both basic, like ammonia, and acidic, like CO2) simultaneously will be key.

@Annu_Nakki @serehfas What's driving the dynamics your sharing could be the result of a lot of factors beyond viral aerostability. It's part of what makes this so challenging, transmission is a complex process and everything matters.

@Whosez5 The reason for this is because humidity affects a lot fo different factors that all affect transmission. For example, room gets humid? People will open a window to feel more comfortable (ventilation). We've measured the viral decay is slowerin cold air, and faster in dry air.

@Whosez5 Your lungs ability to fend off infection is also affected by humidity. Dry air makes it harder for your lungs to stop an infection.