Systematic review with meta-analysis of the accuracy of diagnostic tests for COVID-19

Highlights • RT-PCR followed by CT shows high sensitivity for detecting COVID-19 • Immunological tests should use a combination of IgG and IgM • The genes E and RdRp present high analytical sensitivity to detect the virus • Assays for molecular diagnosis should employ two-target systems • Studies of diagnostic tests for COVID-19 are of moderate methodological quality
Objective . To collate the evidence on the accuracy parameters of all available diagnostic methods for detecting SARS-CoV-2. Methods . A systematic review with meta-analysis was performed. Searches were conducted in Pubmed and Scopus (April 2020). Studies reporting data on sensitivity or specificity of diagnostic tests for COVID-19 using any human biological sample were included. Results . Sixteen studies were evaluated. Meta-analysis showed that computed tomography has high sensitivity (91.9% [89.8–93.7%]), but low specificity (25.1% [21.0–29.5%]). The combination of IgM and IgG antibodies demonstrated promising results for both parameters (84.5% [82.2%-86.6%]; 91.6% [86.0%-95.4%], respectively). For RT-PCR tests, rectal stools/swab, urine, and plasma were less sensitive while sputum (97.2% [90.3–99.7%]) presented higher sensitivity for detecting the virus. Conclusions . RT-PCR remains the gold standard for the diagnosis of COVID-19 in sputum samples. However, the combination of different diagnostic tests is highly recommended to achieve adequate sensitivity and specificity.