Background: Management and control of coronavirus disease 2019 (COVID-19) relies on reliable diagnostic testing.

Objectives: To  evaluate the diagnostic test accuracy (DTA) of nucleic acid  amplification tests (NAATs) for the diagnosis of coronavirus infections.

Data sources: PubMed,  Web of Science, the Cochrane Library, Embase, Open Grey and conference  proceeding until May 2019. PubMed and medRxiv were updated for COVID-19  on 31st August 2020.

Study eligibility: Studies were eligible if they reported on agreement rates between different NAATs using clinical samples.

Participants: Symptomatic patients with suspected upper or lower respiratory tract coronavirus infection.

Methods: The  new NAAT was defined as the index test and the existing NAAT as  reference standard. Data were extracted independently in duplicate. Risk  of bias was assessed using the Quality Assessment of Diagnostic  Accuracy Studies 2 tool. Confidence regions (CRs) surrounding summary  sensitivity/specificity pooled by bivariate meta-analysis are reported.  Heterogeneity was assessed using meta-regression.

Results: Fifty-one  studies were included, 22 of which included 10 181 persons before  COVID-19 and 29 including 8742 persons diagnosed with severe acute  respiratory syndrome coronavirus 2 (SARS-CoV-2). The overall summary  sensitivity was 89.1% (95%CR 84.0-92.7%) and specificity 98.9% (95%CR  98.0-99.4%). Nearly all the studies evaluated different PCRs as both  index and reference standards. Real-time RT PCR assays resulted in  significantly higher sensitivity than other tests. Reference standards  at high risk of bias possibly exaggerated specificity. The pooled  sensitivity and specificity of studies evaluating SARS-COV-2 were 90.4%  (95%CR 83.7-94.5%) and 98.1% (95%CR 95.9-99.2), respectively. SARS-COV-2  studies using samples from the lower respiratory tract, real-time  RT-PCR, and tests targeting the N or S gene or more than one gene showed  higher sensitivity, and assays based on reverse transcriptase  loop-mediated isothermal amplification (RT-LAMP), especially when  targeting only the RNA-dependent RNA polymerase (RdRp) gene, showed  significantly lower sensitivity compared to other studies.

Conclusions: Pooling  all studies to date shows that on average 10% of patients with  coronavirus infections might be missed with PCR tests. Variables  affecting sensitivity and specificity can be used for test selection and  development.

Keywords: Acute respiratory tract infection; COVID-19; Coronavirus; Nucleic acid amplification tests; SARS-CoV-2.