A lateral flow immunoassay test performance in SARS-CoV-2 seroprevalence surveys: a validation study among healthcare workers

Garlantézec, Ronan; Heslan, Christopher; Tadie, Emilie; Tattevin, Pierre; Thibault, Vincent; Paris, Christophe

The objective of this study was to evaluate the validity and reliability of NG-Test® when used as a finger-prick test on healthcare workers and to compare it to the ELISA Wantai Immunoassay. Fifty-one healthcare workers who were RT-PCR SARS-CoV-2 positive and 59 who were RT-PCR SARS-CoV-2 negative accepted to participate in this study. They were subjected to an NG-Test® finger-prick test and collection of a blood sample on the same day. A second NG-Test® on another finger was performed for the first 30 cases and controls and read blinded to the first. Sera obtained from blood samples were used to perform the Wantai SARS-CoV-2 ELISA. The interobserver agreement for the NG-Test® test was perfect (kappa coefficient = 100% [98%–100%]). The sensitivity of NG-Test® was estimated to be 85% [71.9%–92.3%] and the specificity 98.3% [95.0%–100.0%]) for both IgG and IgM. The percentage of agreement between the Wantai immunoassay and NG-Test® was 92.73% for IgG (Kappa = 0.85 [0.75–0.95]) and 65.45% (Kappa = 0.42 [0.26–0.58]) for IgM. Our study highlights the need to validate rapid immunoassay tests under real-life conditions. If NG-Test® is used in seroprevalence surveys, we recommend that its diagnostic performance be taken into consideration to obtain a reliable estimation.