Kanti Pabbaraju, Thomas J. Louie, Linda R. Ward, John Conly, Jia Hu, Ryan S. Noyce, Kevin Fonseca, David H. Evans, Linet Kiplagat, Rebecca J. Malott, Kara Gill, Yi-Chan Lin, and Byron M. Berenger
Few studies have assessed for infectious SARS-CoV-2 in multiple types of clinical and environmental samples. In almost 500 samples from 75 hospitalized and community cases, we detected infectious virus with quantitative burdens varying from 5.0 plaque-forming units/mL (PFU/mL) up to 1.0×106 PFU/mL in clinical specimens and up to 1.3×106 PFU/mL on fomites including facial tissues, nasal prongs, call bells/cell phones, dentures, and sputum deposits with confirmation by plaque morphology, PCR, immunohistochemistry, and sequencing. Expectorated sputum samples had the highest percentage of positive samples and virus titers (71%, 2.9×102 to 5.2×105 PFU/mL), followed by saliva (58%, 10 to 4.6×104 PFU/mL), and cough samples without sputum (19%, 5 to 1.9×103 PFU/mL). We also detected infectious SARS-CoV-2 from patients’ hands (28%, 60 to 2.3×102 PFU/mL) but no infectious virus was found in continuous speech samples despite finding high levels of infectious virus in the associated nasopharynx, throat, or saliva specimens. We demonstrated infectious virus stability in clinical samples, including those dried for prolonged periods of time. Infectious virus correlated with time since symptom onset with no detection after 7-8 days in immunocompetent hosts and with N-gene based Ct values ≤ 25 significantly predictive of yielding plaques in culture. One PFU was associated with ∼105 copies of N gene RNA across a diversity of samples and times from symptom onset. Clinical salivary isolates caused illness in a hamster model with a minimum infectious dose of ≤14 PFU/mL. Our findings of high quantitative burdens of infectious virus, stability even with drying, and a very low minimal infectious dose suggest multiple modes of transmission are exploited by SARS-CoV-2, including direct contact, large respiratory droplet, and fomite transmission and in the context of a high binding avidity to human cellular receptors, offer an explanation of the high contagiousness of this virus.Research in ContextEvidence before this studyWe searched the literature for articles that reported on the presence of infectious SARS-CoV-2 in patients’ samples from clinical and environmental sources. We found several key primary studies and systematic reviews providing valuable background on the carriage of infectious virus and the correlation with cycle threshold (Ct) and/or RNA copies/mL on PCR testing. Clinical correlations with respect to underlying clinical conditions and details on the onset of illness were not commonly reported with respect to the timing of obtaining specimens for culture. Few studies carefully assessed the presence of infectious virus in cough samples, sputum, nasal secretions, hands, and common high touch surfaces. A few published works were found on factors which may be associated with shedding of infectious virus.Added value of this studyWe assessed the presence of infectious virus shedding in almost 500 specimens from 75 patients with COVID-19 in both the hospital and community setting. High titers of infectious virus were detected in multiple clinical and environmental samples. The longest duration of recovery of infectious virus in a fomite sample was from a dried facial tissue found at a patient’s bedside table, used at least 9 hours earlier. Cough specimens revealed infectious virus in 28% of specimens with infectious virus titers as high as 5.2×105 PFU/mL. Hand samples contained infectious virus with titers ranging from 55 to 2.3×102 PFU/mL. Infectious viral loads correlated with N-gene based Ct values and showed that Ct values ≤ 25 were predictive of yielding plaques in culture. These experiments also showed that infectious virus is most often recovered during a 7 to 8-day period following illness onset in immunocompetent persons, and during that time the ratio of RNA/PFU in these clinical specimens varies relatively little, with a ratio ∼160,000:1. Infectious virus may be recovered for weeks to several months in immunosuppressed persons. We also showed that virus recovered from saliva specimens, representing a commonly encountered fomite sample, caused infection in the Syrian hamster model, hence demonstrating the infectiousness of the virus sourced from this type of specimen. A challenge dose as low as 14 PFU/mL yielded infection in this model.Implications of all the available evidenceWe have shown that SARS-CoV-2 is relatively easy to culture when obtained early in the course of illness and there are high levels of cultivatable SARS-CoV-2 in multiple types of clinical specimens and common fomites, including high-touch surfaces and demonstrated their infectiousness in a mammalian host. Our results demonstrate the presence of high quantitative burdens of SARS-CoV-2 in sputum, saliva, and droplets from coughing, which would lend support to large respiratory droplet transmission, hands which would support direct contact transmission, and fomites which would promote indirect contact transmission. We were unable to detect any infectious virus in continuous speech samples which suggests that brief conversations, without coughing or sneezing, pose little risk of transmitting SARS-CoV-2. Our findings provide an explanation for the high contagiousness of this virus and support current public health measures and infection prevention and control guidelines including physical distancing, hand hygiene, masking, and cleaning and disinfection.