After the initial reluctance to recognize the aerosol transmission route as being of any significant importance in Covid-19, the scientific community seems to have settled upon the idea that it has probably been important in certain indoor superspreader events. This paper puts forward additional lines of evidence suggesting that the aerosol route is not just indispensable for explaining such events but is the dominant transmission route of the disease. We also turn on its head the main objection to the aerosol route’s importance, namely, that it would imply much larger spread than what has so far been observed. Building on recent experimental studies on influenza and evidence on Covid-19, we discuss evidence that Covid-19 may is mostly spread by the minority of infected with significant lung involvement, who exhale enough infectious aerosol.
The damage that has already been caused by the novel COVID-19 disease and the non-pharmaceutical measures undertaken by governments in order to slow down its spread makes it urgent to better understand the actual physical mode or modes of transmission. Depending on which mode is predominant, it may be possible to reduce the spread better and potentially achieve this objective with lower impact on the economic and social life.
Before the COVID-19 pandemic, the question whether the large droplet or aerosol (airborne) transmission route is the predominant one had already been a subject of controversy in the research on other viral respiratory diseases, especially influenza.
The two crucial differences between the aforementioned transmission routes lie in the behavior of the respective physical carriers of the virus and their capacity to spread the virus directly to the lower respiratory tract, where the virus can cause a lot more damage if it takes hold. Large droplets are too heavy to remain suspended in the air and follow a ballistic trajectory before either falling to the ground or landing on an object. They are too large to reach the lungs. Aerosols (or droplet nuclei) can continue floating in the air long after their release. Some of them are so small that they can be carried into the lower respiratory tract.
The role of aerosols in the transmission of respiratory viruses had already been a source of controversy before the emergence of COVID-19. The 2013 review of evidence on influenza transmission routes by Killingley and Nguyen-Van-Tam cast doubt on the primacy of the large droplet route in influenza transmission. Among other things, they stressed the finding that it takes a much smaller dose of the virus to infect people using inhaled aerosols than intranasal inoculation. The authors also noted that coupled with the need for the potential target of infection to inhale large droplets at exactly the right time, the higher required infectious dose of large droplets had made some authors cast a strong doubt on their importance. Nonetheless, this view has not yet become mainstream in the epidemiological and virological community, as evidenced by the predominant perspective on Covid-19 spread.
The predominant hypothesis with regard to SARS-CoV-2 spread has so far been that it happens primarily through large droplets. However, recently, Morawska and Milton, with support from multiple other COVID-19 researchers, have called upon the scientific community to recognize the importance of the aerosol route. In its scientific brief issued in response, the World Health Organization has generally disagreed with the signatories of the open letter, only conceding that aerosol-driven spread may occur in very specific circumstances.
So far, the key argument invoked in support of the aerosol route’s importance in Covid-19 has been the certain major superspreader events such as the Skagit Valley chorale and the Diamond Princess cruise ship can hardly be explained without invoking aerosols. The WHO in  agreed that there may have been aerosol spread in those cases but also noted that this was not the only possible explanation.
If major indoor superspreader events were indeed the only line of evidence in favor of the aerosol route’s importance, we would indeed have to conclude that there is currently not enough evidence for it. However, in what follows, we will attempt to demonstrate that there are converging lines of evidence suggesting not only the role of aerosol transmission in Covid-19 is prominent but that this route of transmission is, in fact, the primary one.
Evidence of the aerosol route’s primacy in Covid-19
It seems to be generally agreed at this point that COVID-19 spreads mostly in indoor spaces, even though few studies have been published on outdoor vs indoor transmission. A Chinese study of 318 transmission clusters involving 1245 confirmed cases found only one cluster involving just two people where the infection may have spread outdoors. A study of early Japanese transmission clusters concluded that indoor transmission was 18.7 times more probable than the indoor one. However, it is noteworthy that in the discussion of the actual transmission environments, the authors of the study do not seem to mention any purely outdoor context. In addition to those studies, researchers failed to find evidence that the Black Lives Matter protests in multiple U. S. cities during the pandemic had caused increases in COVID-19 incidence.
The paucity of evidence of any significant outdoor spread of Covid-19 seems difficult to explain if large droplets are considered as the predominant infection route. The way infection would presumably take place in this mode is by the infectee inhaling the droplets ejected a short distance away. It is unclear what environmental differences between outdoor and indoor conditions could render this mode of spread almost inoperative, especially in the contexts of prolonged close interaction such as protests. At the same time, if the aerosol route were primary, very little outdoor spread would be unsurprising.
Another striking piece of evidence in favour of the aerosol route concerns the paucity of COVID-19 spread on airplanes. During a typical flight, multiple passengers spend significant amounts of time tightly packed within an enclosed space. If large droplets were behind most transmission, one would expect airplanes to be an important transmission environment.
However, a recent report by the International Air Transport Association (IATA) cites only one case in which a public health investigation concluded that SARS-CoV-2 had presumably been transmitted from one infected passenger to 12 other nearby-seated passengers and a cabin crew member. In contrast, two other studies mentioned in the report showed that transmission appears to have failed to take place on board of two flights. Most strikingly, if failed to occur during a 31 March flight from the USA to Taipei, despite the presence of 12 passengers who were symptomatic in-flight. The report also cites the results of an informal survey among 18 airlines that report very few incidents of clear onboard spread.
The in-flight context certainly does differ from other close interactions in terms of the potential large-droplet transmission. The IATA report notes that passengers are not oriented face-to-face and only briefly leave their seats. However, if large droplets were the driver of spread, these features of airplanes would seem insufficient to explain the low evidence of in-flight transmission.
At the same time, airplane cabins differ significantly from many other indoor space contexts in terms of the quality of ventilation. The IATA report notes that they are characterized by strong ceiling-to-floor airflow and universal use of HEPA filters. Coupled with the low evidence of in-flight spread, the superior ventilation characteristics of airplane cabins strongly suggest the primacy of the aerosol infection route.
Another boost for the aerosol route’s primacy hypothesis is provided by the studies showing that people infected in contexts presumably facilitating higher aerosol spread tend to have more severe symptoms. A study of the spread of COVID-19 in Gangelt, Germany, found that the participants of the superspreader event that took place in a large hall tended to have more severe symptoms than their infected secondary contacts. Similarly, 36% of the infected during a choir practice at a small non-ventilated indoor space in France required hospitalization and 21% of the infected were admitted to the ICU.
The final line of evidence that started to emerge recently concerns the significantly better protection from Covid-19 infection afforded by high-filtration (aerosol-blocking) respirators compared to other types of face coverings. A study conducted in several Finnish hospitals found that 63.6% of the occupational infections registered among the medical workers studied occurred while using a surgical mask. Meanwhile, none of the medical workers got infected while wearing a FFP2/3 respirator.
The key potential objections to the aerosol route primacy hypothesis
Until recently, the main objection to the hypothesis that the aerosol route is predominant in COVID-19 concerned the lack of evidence that infectious virus could be isolated from the exhaled breath of the infected and the air close to them. However, lately, several studies have addressed this problem by showing that it was indeed possible to isolate the virus that way.
The sceptics of the aerosol primacy hypothesis could also wonder why, if SARS-CoV-2 is primarily spread by people with lung involvement, is it so much more contagious that SARS and MERS that are mostly contained to the lungs? The potential answer may be given by the findings by Sun et al. In particular, they discovered that SARS-CoV-2 infection in Calu-3 lung cells resulted in a much earlier host cellular and immune responses than those by SARS-CoV and MERS-CoV. If lung inflammation is behind infectious aerosol production, the early immune response to SARS-CoV-2 may account for its higher infectiousness compared to SARS and MERS without the need to invoke the upper respiratory tract. People at the early stage of disease with cryptic lung involvement would be able to transmit the virus early, whereas in SARS and MERS, transmission capability would coincide with the highly symptomatic stage of the infection.
The final major possible objection concerns the fact that the spread of Covid-19 has so far been limited even in places like Stockholm where there were a significant initial importation and few government restrictions on movement and contacts. One could argue that if SARS-CoV-2 were anything but only opportunistically airborne,the spread in places like Stockholm would have been much more massive.
This objection, however, could be turned on its head if we hypothesized that only a minority of those infected with SARS-CoV-2 were capable of producing substantial amounts of infectious aerosols. If only a minority of the infected are capable of producing significant amounts of infectious aerosols, then the relatively limited spread of COVID-19 even in places like Stockholm would in fact be easier to explain if the aerosol route were considered as primary.
Recent evidence from influenza
Even before the COVID-19 pandemic, evidence in favour of the hypothesis that only a minority of the infected were infectious because only they could produce significant amounts of infectious aerosol had started to emerge from the research into influenza.
A recent large human challenge study involved two groups of volunteers. The first group (the donors) was inoculated through the nose by the researchers, while the second was supposed to be infected by the donors. Remarkably, despite presumably highly suitable conditions for transmission (a closed space with low humidity), the donors collectively managed to infect only one of their targets.
The researchers noted that the deliberately infected participants did not produce as much of infectious aerosols as symptomatic people in a natural infection setting had been found to. They hypothesized that a minority of the infected with significant lung involvement may be behind most of transmission.
The hypothesis that infectious aerosols in influenza are primarily generated in the lower respiratory tract is supported by the study conducted in naturally infected college students. The key finding was that the presence of infectious virus in fine aerosols exhaled by the infected was not correlated with other respiratory symptoms, such as sneezes and even coughs.
While SARS-CoV-2 is different from influenza viruses, there are reasons to believe that it could behave similarly in terms of its transmission routes. Like influenza, SARS-CoV-2 is a respiratory virus that can reach and infect cells in the lower respiratory tract if it is brought there with aerosols. It potentially faces the same difficulties of infecting the upper respiratory tract while being carried in large droplets, as have been hypothesized for influenza. Its main known differences from influenza viruses concern the mechanism of cell entry and what happens after the viral RNA and proteins have been released into cells but they do not seem to affect the potential primary transmission mode in obvious ways.
Intriguingly, at least in the temporal regions, the efficiency of indoor transmission of SARS-CoV-2 appears to depend on relative humidity in a manner similar to influenza. This finding suggests the existence of important similarities in terms of physical modes of transmission.
Furthermore, although, to our knowledge, there have so far been no human challenge studies in Covid-19 similar to the one described in , evidence has emerged indicating that, just like in influenza, just a minority of the infected could be driving the spread, and the lower respiratory tract involvement could determine the degree of infectiousness by determining the production of infectious fine aerosols.
Evidence that SARS-CoV is spread by a minority of the infected
Several recent studies present evidence of an outsized role of superspreaders in the transmission of COVID-19, including the clues from the viral genomes sequenced worldwide, as well as in Israel and Hong Kong.At the same time, contact tracing studies have generally found low secondary attack rates even within households, even though household contacts tend to spend significant amounts of time in relative proximity to the infected members of the same household.These twin facts strongly suggest that only a minority of the infected are themselves significantly infectious.
It could be argued that this evidence is not sufficient to make this conclusion because there could be something special about the superspread contexts themselves, rather than about the infectors in them, that makes them different from the household context. However, it is unclear what that differentiating factor or factors could be. Some indoor spaces where superspreader events took place may have been more poorly ventilated than average household living spaces and (or) involved loud speech or singing, either of which could result in very high rates of aerosol emission over a concentrated period of time. However, household contacts tend to spend much more time in the proximity of the infected members than a typical person infected during a superspreader event, which would tend to reduce the differences in exposure.
It could also be objected that the apparent outsized role of superspreaders in transmission coupled with low transmission in household may instead be pointing at pre-existing cross-reactive T-cell immunity from other endemic human coronaviruses that has been hypothesized for SARS-CoV-2 based on evidence of cross-reactivity,even though the efficacy of this immunity remains to be proven. We note, however, that in certain contexts such as the State of Maranhao in Brazil,the Austrian village of Ishgland the slum-dominated areas of Mumbai,the seroprevalence findings indicate the spread to at least 40–60% of the population. Such high seroprevalence would suggest that even if cross-immunity could be at work to a certain extent, its impact could well be rather limited.
Evidence that those of the infected who are contagious have significant lung involvement
In addition to the evidence that a minority of the infected have been responsible for the bulk of the spread of COVID-19, research may be increasingly supporting the idea that lung involvement is needed for significant contagiousness. A study of 391 index cases in China found that contagiousness tended to increase significantly with disease severity. Tellingly, the most contagious category of patients were those with expectoration.
In addition to this while those infected with SARS-CoV-2 who have no symptoms appear to be only weakly contagious, they nonetheless tend to have similar nasal viral loads to the symptomatic patients. Similarly, the infected children have been found to have comparable viral loads to those of adults.
At the same time, some studies strongly suggest that severe disease is correlated with higher viral load in exhaled aerosols and the lungs. Moreover, disease severity in  was found to be much better correlated with the lung viral loads than the nasal ones. Moreover, studies in children and adolescents have shown low levels of disease severity and lung involvement.
In addition to this, some of the patients who are asymptomatic at the time of testing may actually have significant cryptic lung involvement. For instance, in one study among the passengers of the “Diamond Princess” cruise ship, 54% of the asymptomatic patients’ CT scans analysed had detectable lung abnormalities. This may explain how some asymptomatic patients could be contagious. The high proportion of asymptomatic patients with lung scan abnormalities is probably explained by the high average age of cruise ship passengers and the close-quarter conditions on the ship.
Evidence for the role of fine aerosols
A potential objection to the hypothesis that only a minority with lung involvement are capable of producing significant amounts of infectious aerosols concerns the possibility of aerosols being generated in the upper respiratory tract of the infected.
In the study discussed in , the replication of SARS-CoV-2 could only be observed to a statistically significant degree in the samples from the air exhaled by the infected in the aerosol particles of the smallest size, or fine aerosols. This strongly suggests that infectious aerosols are primarily produced in the lower respiratory tract, more specifically, the alveoli — just like the study authors in  proposed in influenza.
Furthermore, COVID-19 infection has been found to induce exactly the kind of severe alveolar inflammation that would be necessary to produce the infectious fine aerosols.
This paper has presented and discussed the emerging evidence pointing at aerosol-driven spread by a minority of the infected with significant lung involvement being at the core of COVID-19 transmission.
If its conclusions are found to be correct, the main implication will be that more focus may be needed on better indoor ventilation and relative humidity management compared to physical distancing and the wearing of face coverings with respect to mitigating the COVID-19 damage. This would also underscore once more the potential of controlling COVID-19 outbreaks by containing the spread driven by indoor superspreader events.
 Killingley, B. & Nguyen-Van-Tam, J. S. Routes of influenza transmission. Influenza and other respiratory viruses, 2013; 7(2): 42–51.
 Morawska, L. & Milton, D.K. It is Time to Address Airborne Transmission of COVID-19. Clinical Infectious Diseases, 2020; ciaa939.
 World Health Organization. Transmission of SARS-CoV-2: implications for infection prevention precautions: Scientific Brief. WHO, 2020; doi: https://www.who.int/news-room/commentaries/detail/transmission-of-sars-cov-2-implications-for-infection-prevention-precautions (accessed 05/09/2020).
 Miller S. L. et al. Transmission of SARS-CoV-2 by inhalation of respiratory aerosol in the Skagit Valley Chorale superspreading event. medRxiv, 2020; doi: https://doi.org/10.1101/2020.06.15.20132027 (accessed 05/09/2020).
 Almilaji, O., & Thomas, P. Air recirculation role in the infection with COVID-19, lessons learned from Diamond Princess cruise ship. medRxiv, 2020; doi: https://doi.org/10.1101/2020.07.08.20148775 (accessed 05/09/2020).
 Nishiura, H. et al. Closed environments facilitate secondary transmission of coronavirus disease 2019 (COVID-19). medRxiv, 2020; doi: https://doi.org/10.1101/2020.02.28.20029272 (accessed 05/09/2020).
 Dave, D. M. et al. Black Lives Matter Protests, Social Distancing, and COVID-19. The National Bureau of Economic Research, 2020; Working Paper №27408.
 International Air Transport Association. Restarting aviation following COVID-19: Medical evidence for various strategies being discussed as at 09 June 2020 IATA Medical Advisory Group. IATA, 2020; doi: https://www.iata.org/contentassets/f1163430bba94512a583eb6d6b24aa56/covid-medical-evidence-for-strategies-200609.pdf (accessed 05/09/2020).
 Streeck, H. et al. Infection fatality rate of SARS-CoV-2 infection in a German community with a super-spreading event. medRxiv, 2020; doi: https://doi.org/10.1101/2020.05.04.20090076 (accessed 05/09/2020).
 Charlotte, N. High Rate of SARS-CoV-2 Transmission due to Choir Practice in France at the Beginning of the COVID-19 Pandemic. medRxiv, 2020; doi: https://doi.org/10.1101/2020.07.19.20145326 (accessed 05/09/2020).
 Oksanen L-M. et al. Healthcare workers high COVID-19 infection rate: the source of infections and potential for respirators and surgical masks to reduce occupational infections. medRxiv, 2020; doi: https://doi.org/10.1101/2020.08.17.20176842 (accessed 05/09/2020).
 Santarpia, J. L. et al. Aerosol and surface contamination of SARS-CoV-2 observed in quarantine and isolation care. Scientific Reports, 2020; 10(12732).
 Sun, J. et al. Comparative transcriptome analysis reveals the intensive early-stage responses of host cells to SARS-CoV-2 infection. bioRxiv, 2020; doi: https://doi.org/10.1101/2020.04.30.071274 (accessed 05/09/2020).
 Orlowski, E. J. W. & Goldsmith, D. J. A. Four months into the COVID-19 pandemic, Sweden’s prized herd immunity is nowhere in sight. Journal of the Royal Society of Medicine, 2020; 113(8), 292–8.
 Hota, B. et al. Estimate of airborne transmission of SARS-CoV-2 using real time tracking of health care workers. medRxiv, 2020; doi: https://doi.org/10.1101/2020.07.15.20154567 (accessed 05/09/2020).
 Nguyen-Van-Tam, J. S. et al. Minimal transmission in an influenza A (H3N2) human challenge-transmission model with exposure events in a controlled environment. PLOS Pathogens, 2020; 16(7): e1008704.
 Yan, J. et al. Infectious virus in exhaled breath of symptomatic seasonal influenza cases from a college community. Proceedings of the National Academy of Sciences, 2018; 115(5): 1081–6.
 Ahlawat, A., Wiedensohler, A. and Kumar Mishra, S. An Overview on the Role of Relative Humidity in Airborne Transmission of SARS-CoV-2 in Indoor Environments. Aerosol and Air Quality Research, 2020; 20(9).
 Lowen, A. C. & Steel, J. Roles of Humidity and Temperature in Shaping Influenza Seasonality. Journal of Virology, 2014; 88 (14): 7692–5.
 Adam, D. et al. Clustering and superspreading potential of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections in Hong Kong. Research Square, 2020; doi: https://doi.org/10.21203/rs.3.rs-29548/v1 (accessed 05/09/2020).
 Madewell, Z. J. et al. Household transmission of SARS-CoV-2: a systematic review and meta-analysis of secondary attack rate. medRxiv, 2020; doi: https://doi.org/10.1101/2020.07.29.20164590 (accessed 05/09/2020).
 Grifoni, A. et al. Targets of T Cell Responses to SARS-CoV-2 Coronavirus in Humans with COVID-19 Disease and Unexposed Individuals. Cell, 2020; 181(7): 1489–1501.
 Moura da Silva, A. A. et al. Population-based seroprevalence of SARS-CoV-2 is more than halfway through the herd immunity threshold in the State of Maranhao, Brazil. medRxiv, 2020; doi: https://doi.org/10.1101/2020.08.28.20180463 (accessed 05/09/2020).
 Malani, A. et al. Seroprevalence of SARS-CoV-2 in slums and non-slums of Mumbai, India, during June 29-July 19, 2020. medRxiv, 2020; doi: https://doi.org/10.1101/2020.08.27.20182741 (accessed 05/09/2020).
 Luo, L. et al. Contact Settings and Risk for Transmission in 3410 Close Contacts of Patients With COVID-19 in Guangzhou, China. Annals of Internal Medicine, 2020; doi: https://doi.org/10.7326/M20-2671
 Lee, S. et al. Shedding Among Asymptomatic and Symptomatic Patients With SARS-CoV-2 Infection in a Community Treatment Center in the Republic of Korea. JAMA Internal Medicine. Published online August 06, 2020; doi: https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2769235
 Tan, J. et al. Transmission and clinical characteristics of asymptomatic patients with SARS-CoV-2 infection. Future Virology, 2020; 15(6).
 Baggio, S. et al. SARS-CoV-2 viral load in the upper respiratory tract of children and adults with early acute COVID-19. Clinical Infectious Diseases, 2020; ciaa1157.
 Williams, C. M. et al. COVID-19: Exhaled virus detected by Face Mask Sampling provides new links to disease severity and potential infectivity. medRxiv, 2020; doi: https://doi.org/10.1101/2020.08.18.20176693 (accessed 05/09/2020).
 Hamed, I. et al. Paired nasopharyngeal and deep lung testing for SARS-CoV2 reveals a viral gradient in critically ill patients: a multi-centre study. medRxiv, 2020; doi: https://www.medrxiv.org/content/10.1101/2020.07.19.20156869v1
 RIVM. Children and COVID-19. RIVM, 2020; doi: https://www.rivm.nl/en/novel-coronavirus-covid-19/children-and-covid-19 (accessed 08/07/2020).
 Zhang, C. et al. Clinical and epidemiological characteristics of pediatric SARS-CoV-2 infections in China: A multicenter case series. PLoS Medicine, 2020; 17(6), e1003130.
 Götzinger, F. et al. COVID-19 in children and adolescents in Europe: a multinational, multicentre cohort study. The Lancet Child and Adolescent Health, 2020; S2352–4642(20)30177–2.
Inui, S. et al. Chest CT Findings in Cases from the Cruise Ship “Diamond Princess” with Coronavirus Disease 2019 (COVID-19). Radiology: Cardiothoracic Imaging, 2020; 2(2).