The Role and Diagnostic Accuracy of Serology for COVID-19

2021 
Background: The role and performance of various serological tests for the diagnosis of COVID-19 are unclear. This study aimed to evaluate the performance of seven commercially available serological assays for SARS-CoV-2 antibodies by testing COVID-19 cases and controls. Methods: Adults with fever for more than five days with or without acute respiratory symptoms, admitted to a tertiary-care teaching hospital in South India, were enrolled prospectively between June and December 2020. Cases included SARS-CoV-2 RT-PCR confirmed patients. Controls included those with laboratory-confirmed alternative diagnosis and healthy participants. All participants were tested for IgM, IgG, and total antibodies using ELISA and CLIA platforms. Sensitivities, specificities, PPV, NPV, and accuracies were estimated. Findings: Among 303 participants (153 cases and 150 controls) ELISA detecting anti-S protein antibody had sensitivity of 86.3% and 88.9%; and specificity of 76.7% and 82% for IgM and IgG, respectively. ELISA (85%) and CLIA (80-85%) had similar diagnostic accuracies for antibody detection. Antibody test positivity progressively increased with duration of illness, reaching more than 90% after day 10. On comparing cases against pre-pandemic controls, IgG had excellent specificity (97.6%), sensitivity (88.9%) and accuracy (90.8%), hence would be useful for seroprevalence studies. Interpretation: Serological assays are important adjuncts for diagnosing COVID-19, especially beyond 10 days of illness and for seroprevalence studies. ELISA and CLIA platforms showed similar diagnostic accuracies. Funding Information: Wellcome Trust/DBT India Alliance fellowship (IA/CPHS/16/1/502679) to GMV and Christian Medical College, Vellore. Declaration of Interests: The Authors declare no conflict of interest. Ethics Approval Statement: This study was approved by the institutional review board and ethics committee (No.13166/22.07.2020).
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []