Viral Shedding and COVID-19 Superspreading Events

A Review by the Fellows Behind the Literature Updates

Eric A. Meyerowitz, MD; Aaron G. Richterman, MD, MPH


June 08, 2020

Editor's note: Find the latest COVID-19 news and guidance in Medscape's Coronavirus Resource Center.

There is mounting evidence that SARS-CoV2 has substantial transmission heterogeneity and that superspreading events have driven much of the local epidemics with consequent implications for mitigation strategies.

When we talk about spread of a virus in general, the basic reproductive number, or R0, is important to understand. This is a measure of the average number of secondary infections from an index case in an entirely susceptible population. For an outbreak to persist, the R0 must be greater than 1; if it is less than 1, the number of new infections over time will decrease.

With uniform transmission and an R0 of 2, for instance, each index case transmits to 2 new cases, which then transmit to 2 new cases and so on. For viruses with significant transmission heterogeneity, most cases lead to 0 secondary cases, but a small proportion infect the vast majority of secondary cases. A single instance associated with a large number of secondary transmissions is called a "superspreading event."

Overdispersion and "Superspreading Events"

Heterogenous transmission. The index case leads to 2 new cases, 1 of which yields no additional infections, but the other yields 8 additional cases.

The factors that contribute to superspreading are not completely understood and and are likely multiple, including biological factors (features possibly related to viral load), behavioral and social factors (certain people have a high number of contacts relative to the general population), high-risk facilities (e.g., closed spaces with poor ventilation), and what are called "opportunistic" situations (large events with lots of contacts, such as a wedding), or temporary increases in transmissibility (as occurs with loud talking or singing).

The evidence that superspreading events are important in SARS-CoV2 includes early modeling data looking at local/imported cases, extensive contact tracing in Hong Kong, and modeling based on viral genomes in Israel.

There are now multiple reports of superspreading events from across the globe:

Layout of office floor of call center. Blue = infected.

  • At a wedding in Jordan,  76 of 350 wedding guests developed COVID-19 and 1 died, with the index patient thought to be the father of the bride.

  • At a church in Arkansas, 2 symptomatic people who attended services over the span of 3 days led to 35 of 92 attendees acquiring COVID-19, 3 of whom died.

Date of symptom onset among churchgoers with laboratory-confirmed cases of COVID-19 (n = 35) who attended services between March 6 and 11.

  • At a Washington choir practice, 1 symptomatic person is thought to have led to 52 likely infections, or 87% of the choir, with 2 subsequent deaths.

Confirmed and probable cases of COVID-19 associated with 2 choir practices, by date of symptom onset.

Hospital Spread and the Role of Fomites

A remarkable report tracked in detail a hospital superspreading event from St. Augustine's Hospital in Durban, South Africa. The outbreak infected 34 patients (15 of whom died) and 80 staff members across 5 different hospital wards and 2 outside facilities (a nursing home and a dialysis center).

Because of the distribution of cases within the wards, the authors hypothesized that indirect contact transmission, through fomites such as stethoscopes, might have played an important role. However, given the likely greater importance of droplet transmission and the high rates of healthcare worker infections, we suspect that infected healthcare workers (asymptomatic or symptomatic) making their way around the wards may have contributed more to the distribution of COVID-19.

The report does not include information on the ventilation of the building, which is an important factor for both droplet and potentially airborne transmission. In healthcare settings with large burdens of endemic tuberculosis, such as South Africa, ventilation is typically emphasized.

Infectiousness, Testing, and the Myth of Reinfection

Polymerase chain reaction (PCR) assays that look for viral genetic material show that some individuals are positive for SARS-CoV2, for weeks or even months. A recent modeling study published in Emerging Infectious Diseases estimated that 50% of infected people are PCR positive until day 22, and some are positive out to 2 months after symptom onset.

However, PCR assays just detect presence of viral RNA and do not measure whether that viral material can cause infection. Until recently it was not clear whether those who are PCR positive for weeks are still infectious.

Now multiple studies  have shown that infectiousness declines rapidly after symptom onset  and, in general, infectious virus has not been isolated 8 days after symptom onset.

One important implication is that "surveillance" PCR testing in known positive individuals has no utility for determining whether a healthcare worker can safely return to work, for example. Instead, it may be useful for labs to report "cycle threshold" as a proxy for viral load. The cycle threshold refers to the number of PCR amplification cycles it takes to identify a gene target and get a positive test result: With more virus, it takes fewer cycles, so a low cycle threshold indicates a high viral load and a high cycle threshold indicates a low viral load. A recent study found no positive viral cultures when the cycle threshold was over 24: This means that if it took more than 24 cycles of amplification to get a positive test result, then no live virus could be cultured.

There are some reports of individuals possibly being reinfected with the virus, according to PCR assays. However, a recent study by the Korean CDC examined 208 people who tested positive after discharge from isolation from 7 to 82 days after symptom onset. Samples from these patients yielded no positive viral cultures, indicating that the material detected was not infectious virus. Additionally, the researchers did extensive contract tracing and found no evidence that these individuals with late PCR positivity had transmitted the virus to any others.

This suggests that early reinfection is a myth and that people with SARS-CoV2 infection do not have a long duration of infectivity. Instead, isolated superspreading events may contribute to most secondary transmission and occur because of a combination of factors, possibly including the high viral load of the index case at the time of the event, the type of event (e.g., an event with singing, during which virus may be aerosolized), and the adequacy of ventilation.

Cutting the Tail: Containment of Spread

Early evidence points to superspreading events of SARS-CoV2 occurring predominantly in closed environments with poor ventilation where people are closely packed and are exposed for long durations, particularly with face-to-face contact (>10 minutes).

The implications for "cutting the tail" of infection spread are many. Viral load is thought to play a major role in superspreading events, and viral load peaks just before symptom onset, making universal masking critically important  to slow the spread of the virus.

Coronavirus copies per nasal and throat swab samples (first 2 panels). Respiratory and aerosol droplets collected for 30 minutes while not wearing or wearing a surgical face mask (right 4 panels).

Because most transmission occurs indoors in poorly ventilated settings, social distancing and gathering outdoors rather than indoors will help decrease the chance of a superspreading event.

Certain venues are thought to be particularly high risk for spread. These include religious venues (particularly with singing), congregate settings (dormitories, elder homes), healthcare settings, and cruise ships, in addition to schools, bars, shops, and conferences. Healthcare settings can dramatically reduce spread through universal masking, social distancing, and use of personal protective equipment. Other settings may need to use similar strategies and avoid certain high-risk activities, like choir singing.

The growing understanding about the limited duration of infectiousness and the importance of superspreading events provides some much-needed good news for control of the pandemic. Outside of these specific superspreading events,  the secondary attack rate is relatively low (<20% for household contacts in most studies). By taking away the opportunity for superspreading events through many of the mitigation strategies described here, we may be able to make a substantial impact on the propagation of the virus without wholesale "lockdown" strategies that come with profound social and economic consequences.

Eric Meyerowitz, MD, from Massachusetts General Hospital and Aaron Richterman, MD, MPH, from Brigham and Women's Hospital, are HIV fellows who collaborate on a periodic deep dive into the COVID-19 literature. When not poring over scientific studies, Eric enjoys spending time with family and friends and Aaron can be found admiring the trees in the Arnold Arboretum in Boston.

You can follow them on Twitter: @EricMeyerowitz  and @AaronRichterman

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.