Diagnostic Testing for COVID-19: Systematic Review of Meta-Analyses and Evidence-Based Algorithms

Abstract

Background: Coronavirus Disease-19 (COVID-19), a disease caused by infection with the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), is a global pandemic. Diagnosis is critical and diagnostic techniques include reverse transcriptase polymerase chain reaction (RT-PCR), serologic antibody testing, and chest computed tomography (CT). Despite rigorous meta-analyses looking into these techniques, there is no summary and additionally no algorithm to help guide diagnostic testing. Our objective is to perform a systematic review of the literature and to provide evidence-based algorithms for diagnosing or ruling out COVID-19.

Methods: Data were gathered using PubMed and Ovid research databases using a predefined medical subject heading (MeSH) based search, and sources that were included in the study had their references reviewed to screen for more sources for this study. Sources were collected up to 23 August 2020. Two researchers searched through the databases for articles and data/articles meeting inclusion criteria were extracted.

Results: 395 articles were identified, and 10 studies were included. Meta-analyses of diagnostic tests were included in our systematic review. An overview was then provided for each diagnostic test. Sensitivities and specificities for RT-PCR, serologic antibody tests and chest CT were collected, and the data was stratified by categorical variables. Two evidence-based algorithms were developed for symptomatic and asymptomatic patients in the hospitalized and ambulatory settings.

Conclusions: This article provides a summary of the up-to-date efficacy of the most utilized diagnostic tests currently available for COVID-19. Additionally, this article provides evidence-based COVID-19 diagnostic algorithms for symptomatic and asymptomatic patients in the hospitalized and ambulatory settings.

Keywords: COVID-19; RT-PCR; antibody; chest CT; meta-analysis.