Clinical performance of oral sponge sampling for detection by RT-PCR of SARS-CoV-2

MARQUETTE, C. H.; Boutros, J.; Benzaquen, J.; Selva, E.; Labaky, M.; Benchetrit, D.; Lavrut, T.; Leroy, S.; Chemla, R.; Carles, M.; Tanga, V.; Maniel, C.; Bordone, O.; Allegra, M.; Lespinet, V.; Fayada, J.; Griffonnet, J.; Hofman, V.; Hofman, P.

Background: The current standard for coronavirus 2019 disease (COVID-19) diagnosis is reverse transcriptase-polymerase chain reaction (RT-PCR) testing of naso-pharyngeal swabs (NPS), Sampling with NPS is invasive and requires specialized and trained personnel, which limits rapid and repeated screening for the disease. A less invasive and possibly self-administered sampling method may increase the capacity for testing and be more effective in identifying, isolating, and filtering out currently infected persons. Methods: Over a period of three months, we included volunteers presenting with recent symptoms suggestive of a SARS-CoV-2 infection at a free COVID-19 screening center in the city of Nice, France. NPS as well as nasal and oral sponges were collected in parallel and analyzed by RT-PCR for SARS-CoV-2. Results: One hundred and forty-seven subjects were included, of whom, 41.5% were diag-nosed with COVID-19 using NPS RT-PCR. RT-PCR on nasal and oral sponges showed a sensitivity of 87 to 98% and 72 to 87%, respectively for diagnosis of COVID-19 in symptomat-ic subjects, depending on the type of RT-PCR technique used. The specificity was 100% whatever the RT-PCR test. The viral load determined with the oral samples was significantly lower than with NPS. Conclusion: Taken together, these results demonstrated that the oral sponge sampling method can be standardized, is easy to use and cheap. The acceptability makes it a repeata-ble test, notably for elderly people or children. It may become a high-frequency - low analyti-cal sensitive testing strategy.