Background: Respiratory infections are a major contributor to hospital admissions. Identification of respiratory pathogens by means of conventional culture and serology methods remains challenging. Multiplex molecular assays are an appealing alternative that endeavours to be rapid, more accurate and less arduous.
Objective: The study aimed to compare the clinical performance of three commercial multiplex molecular assays for respiratory viruses.
Methods: Forty-eight respiratory specimens obtained from patients at Tygerberg Hospital in the Western Cape province of South Africa were studied. These specimens were collected between May 2020 and August 2020. The results of the Seegene Anyplex™ II RV16, FilmArray® Respiratory 2.1 plus Panel (FARP), and QIAstat-Dx® Respiratory SARS-CoV-2 Panel (QRP) were analysed based on the overlapping targets. A composite reference standard was applied to provide a standard reference for comparison.
Results: The overall sensitivity of the Seegene Anyplex™ II RV16 was 96.6% (57/59), the FARP 98.2% (56/57) and the QRP 80.7% (46/57). The overall specificities were 99.8% (660/661), 99.0% (704/711) and 99.7% (709/711), respectively. The QRP failed to detect coronaviruses and parainfluenza viruses in 41.7% (5/12) and 28.6% (4/14) of positive specimens, respectively, while the FARP produced the lowest target specificity of 88.4% (38/43) for rhinovirus/enterovirus.
Conclusion: The overall specificity of all three platforms was comparable; however, the sensitivity of the QRP was inferior to that of the ARV and FARP.
What this study adds: This study adds to the body of performance characteristics described for respiratory multiplex panels, especially in the African context where molecular diagnostics for infectious diseases are gaining momentum.
Keywords: Anyplex™ II RV16; FilmArray® Respiratory 2.1 plus Panel; QIAstat-Dx® Respiratory SARS-CoV-2 Panel; composite reference standard; molecular panels; multiplex reverse transcription polymerase chain reaction; respiratory viruses.
© 2024. The Authors.