Diagnostic accuracy of rapid antigen tests in pre-/asymptomatic close contacts of individuals with a confirmed SARS-CoV-2 infection

BackgroundPre-/asymptomatic close contacts of SARS-CoV-2 infected individuals were tested at day 5 after contact by real-time reverse transcriptase polymerase chain reaction (RT-PCR). Diagnostic accuracy of antigen-detecting rapid diagnostic tests (Ag-RDT) in pre-/asymptomatic close contacts was up till now unknown.MethodsWe performed a prospective cross-sectional diagnostic test accuracy study. Close contacts (e.g. selected via the test-and-trace program or contact tracing app) aged ≥16 years and asymptomatic when requesting a test, were included consecutively and tested at day 5 at four Dutch public health service test sites. We evaluated two Ag-RDTs (BD Veritor™ System Ag-RDT (BD), and Roche/SD Biosensor Ag-RDT (SD-B)) with RT-PCR as the reference standard. Virus culture was performed in RT-PCR positive individuals to determine the viral load cut-off above which 95% was culture positive, as a proxy of infectiousness.ResultsOf 2,678 BD-tested individuals, 233 (8.7%) were RT-PCR positive and BD detected 149 (sensitivity 63.9%; 95% confidence interval 57.4%-70.1%). Out of 1,596 SD-B-tested individuals, 132 (8.3%) were RT-PCR positive and SD-B detected 83 (sensitivity 62.9%; 54.0%-71.1%). When applying an infectiousness viral load cut-off ≥ 5.2 log10 gene copies/mL, the sensitivity was 90.1% (84.2%-94.4%) for BD, 86.8% (78.1% to 93.0%) for SD-B overall, and 88.1% (80.5%-93.5%) for BD, 85.1% (74.3%-92.6%) for SD-B for those still asymptomatic at the actual time of sampling. Specificity was >99% for both Ag-RDTs in all analyses.ConclusionsThe sensitivity for detecting SARS-CoV-2 of both Ag-RDTs in pre-/asymptomatic close contacts is over 60%, increasing to over 85% after applying an infectiousness viral load cut-off.Trial registration numberNot applicable. A study protocol is available upon request.