Seropositivity rate and diagnostic accuracy of serological tests in 2019-nCoV cases: a pooled analysis of individual studies

Abstract

Objective: Currently, detection of SARS-CoV-2 RNA is standard in the diagnosis of COVID-19 (2019-nCoV). However, reliable and rapid serological diagnostic methods to screen SARS-CoV-2 infected patients, including those who do not have overt symptoms, are urgently needed. Most studies have described serological tests based on the detection of SARS-CoV-2-specific IgM and IgG. Here, we attempted to systematically analyze the positive rates and comprehensive diagnostic efficacy of IgM and IgG in response to SARS-CoV-2 infection.

Materials and methods: By systematically searching PubMed, medRxiv, bioRxiv and other databases, studies regarding the detection of peripheral blood IgM and/or IgG related to SARS-CoV-2 were collected. The positive rate, sensitivity (SEN), specificity (SPE), area under the curve (AUC) and corresponding 95% CIs were obtained by weighted quantitative mergence, and the source of heterogeneity was explored by performing a subgroup study and sensitivity analysis.

Results: A total of 30 studies were included, which were comprised of 3856 confirmed SARS-CoV-2 RNA positive cases, 368 suspected RNA negative cases, 1167 asymptomatic carriers, and 2526 RNA negative controls. The corresponding meta-analysis showed that in confirmed cases with 2019-nCoV, the positive rates of single IgM, single IgG and their joint detection related to SARS-CoV-2 were 61.2% (95% CI: 53.4%-69.0%), 58.8% (95% CI: 49.6%-68.0%) and 62.1% (52.7%-71.4%), respectively. In suspected RNA negative cases, the positive rates of single IgM, single IgG and their joint detection were 29.0% (95% CI: 14.0%-44.0%), 37.0% (95% CI: 20.0%-55.0%) and 55.0% (95% CI: 19.0%-90.0%), respectively. Interestingly, IgM/IgG detection also demonstrated a positive rate of 19% (95% CI: 10.0%-27.0%) in asymptomatic cases. Using RT-PCR test as reference, the AUCs of IgM, IgG and IgM/IgG in the diagnosis of 2019-nCoV infection were 0.9656, 0.9766, and 0.9838, respectively. The stratified analyses showed that among confirmed cases with 2019-nCoV, the positive rates of IgM and IgG were 27.3% (95%CI: 19.8%-34.8%) and 22.3% (95% CI: 11.3%-33.3%), respectively, 0-7days following the onset of symptoms, whereas the positive rate of parallel IgM/IgG testing attained 39.3% (95% CI: 24.2%-54.4%). Moreover, the efficacy of antibody testing based on CLIA (chemiluminescence enzyme immunoassays) in diagnosing 2019-nCoV infection was higher than that of LFIA (lateral flow immunoassays) and ELISA (enzyme linked immunosorbent assay).

Conclusions: IgM, IgG and their joint testing exhibited high clinical value in the diagnosis of 2019-nCoV, which may assist in making up for the deficiency of throat swab RNA tests.