ABSTRACT
Background The use of rapid antigen diagnostics tests (Ag-RDT) has gained widespread acceptance as an alternative method for diagnosis of COVID-19 outside of health care settings. Various authors have reported that saliva is a reliable specimen, alternative to nasopharyngeal and mid-nasal swabs, to detect SARS-CoV-2 infections by RT-PCR. We assessed the performance of buccal swabs containing saliva for SARS-CoV-2 detection by Ag-RDT, using mid-nasal specimens as a reference in the northern area of Barcelona (Catalonia, Spain) Methods In the context of routine clinical diagnosis of mild COVID-19 patients, we enrolled 300 adults in a study to directly compare mid-nasal swabs and saliva specimens for SARS-CoV-2 detection by Ag-RDT. When mid-nasal and buccal Ag-RDTs showed discordant results, a third mid-nasal swab was collected and analysed by RT-PCR. Results Paired samples were successfully obtained in 300 suspected cases of SARS-CoV-2 infection. Of the 300 paired samples, Ag-RDT with the mid-nasal swab detected 139 (46.3%) positive COVID-19 cases. In comparison, buccal swabs showed a sensitivity and specificity of 31.7% (44/139) and 98.8% (159/161), respectively. 65 discordant results with positive mid-nasal swabs and negative buccal swabs were tested by RT-qPCR. All samples tested by Rt-PCR resulted positive, with a mean cycle threshold (Ct) of 28.3 (SD 7.3). Conclusion Our findings show that mid-nasal swabs have better performance than buccal swabs for detecting SARS-CoV-2 with Ag-RDT tests. Of note, the sensitivity of buccal samples was affected in samples with high viral loads (Ct<33), suggesting that buccal swabs might not be sensitive enough to detect individuals at risk of transmission. Taken together, the existing literature and the results provided in our analysis we advise against the use of buccal specimens for SARS-CoV-2 diagnostics with Ag-RDT.