False-positive colloidal gold-based immunochromatographic strip assay reactions for antibodies to SARS-CoV-2 in patients with autoimmune diseases

Background The outbreak of the novel 2019 coronavirus disease (COVID-19) was declared a global pandemic by the World Health Organization (WHO) on March 11, 2020. The diagnosis of COVID-19 is frequently based on a positive serological test. We noted the occurrence of false-positive results for COVID-19 in the colloidal gold-based immunochromatographic strip (ICS) assay in sera from patients with autoimmune diseases (ADs). This study aimed to evaluate the possible reasons for the false-positive results in two ICS assays (Wondfo ICS and Innovita ICS) and to investigate the effect of urea dissociation in reducing false-positive results. Methods The sera of 135 patients with ADs, 13 confirmed COVID-19 patients, 95 disease controls, and 120 healthy controls were tested for immunoglobin M (IgM) and IgG against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) using Wondfo and Innovita ICS kits. The distributions of auto-antibodies in antibody-positive and antibody-negative groups were also compared, and bivariable logistic regression was used to assess auto-antibodies associated with false-positive results. A urea dissociation test of ICS was performed for the SARS-CoV-2 antibody-positive samples. Results Specificity of Wondfo ICS for the 95 disease controls was 94.74% compared to 98.95% and 96.84% for Innovita SARS-CoV-2 IgM and IgG, respectively. Specificity of Wondfo ICS for the 120 healthy controls was 97.5% compared to 100% and 99.17% for Innovita SARS-CoV-2 IgM and IgG, respectively. Specificity of Wondfo ICS for AD patients was 73.33% compared to 97.78% and 96.30% for Innovita SARS-CoV-2 IgM and IgG, respectively. Sensitivity was 74.07% for Wondfo compared to 70.37% for Innovita IgM and 66.67% for Innovita IgG. Using the Wondfo ICS, the percentage of elevated rheumatoid factor (RF) level (>20 IU/mL) was higher in the SARS-CoV-2 antibody-positive group compared with the antibody-negative group [27/36 (75.0%) vs. 34/99 (34.34%), P=0.001]. The elevated RF was associated with antibody positivity, with an odds ratio of 4.671 [95% confidence interval (CI), 1.88-11.69]. The specificity of the Wondfo ICS assay for the AD patients was increased from 73.33% to 94.07% after the urea dissociation assay. Conclusions An elevated serum RF level could lead to false-positive results when detecting SARS-CoV-2 antibodies using the Wondfo ICS kit, and the urea dissociation assay would be helpful in reducing the incidence of false-positive results.