Background: Reverse transcription polymerase chain reaction (RT-PCR) is the current standard of reference in the diagnosis of SARS-CoV-2 infection. In outpatient clinical practice, nasopharyngeal swab RT-PCR testing is still the most common procedure. The purpose of this systematic review and meta-analysis was to evaluate the sensitivity of RT-PCR nasopharyngeal assays.

Methods: We searched three databases, including PubMed/MEDLINE, EMBASE, and Cochrane Library, using a comprehensive strategy. Studies investigating the sensitivity of SARS-CoV-2 RT-PCR nasopharyngeal assays in adults were included. Two reviewers extracted data and assessed trial quality independently. Pooled sensitivity and its confidence interval were computed using the meta package in R.

Results: Thirteen studies were found eligible for the inclusion in the systematic review. Out of these, 25 different sub-studies were identified and included in the meta-analysis, which reported the sensitivities of 25 different nasopharyngeal RT-PCR assays. Finally, the overall pooled sensitivity resulted 89% (95% CI, 85.4 to 91.8%).

Conclusion: Our study suggests that RT-PCR assays on nasopharyngeal specimens have a substantial sensitivity for diagnosing SARS-CoV-2 infection.