Respiratory tract infection is a common cause of visits to the hospital emergency department (ED). Rapid and sensitive sample-to-result multiplex nucleic acid amplification tests (NAATs) have emerged which target multiple respiratory pathogens in a single test. Conversely, conventional NAATs have improved throughput and broadened the number of clinically relevant targets available.
Here we (a) compare BioFire FilmArray Respiratory Panel 2.1 plus (RP; 22 targets) to SpeeDx PlexPCR (Plex; 8 targets) for the detection of pathogenic respiratory viruses (b) retrospectively analyse turnaround time (TAT) in the clinical context.
Upper respiratory tract swabs were tested prospectively with RP as part of routine testing (n=1915). Samples were retrospectively tested with Plex with discordant analysis by retesting both RP, Plex and GeneXpert. The overall agreement between RP and Plex was 98.4% (95% CI 97.7-98.9%). Contingency tables for all Plex targets and lower-limit detection studies were performed with comparable results.
The median TAT for RP was 1.78h without clinically significant day-to-day variation. ED collected 76% of samples. Pathogen detection rate was higher in ED samples (30%) compared to inpatient wards (17%) although SARS-CoV-2 detection was lower in ED (5%) than inpatients (9%). For ED samples, 89.3% of results were available prior to discharge, including 71.5% of those with an ED visit time under 4h, and increasing to 92.5% in positive samples.
The RP assay is well-suited for rapid respiratory testing in the context of a laboratory operating 24h and servicing a high-turnover ED. Consistently providing TAT of under 2h has clear potential to facilitate appropriate anti-viral prescription and inform bed allocation. We also demonstrate a high degree of concordance comparing RP to Plex. The RP assay has an advantage with a broader range of targets and rapid sample-to-result TAT, whereas the Plex has an advantage in terms of throughput and cost.
We envisage a combination of RP and Plex that could provide an effective diagnostic strategy to meet the clinical and laboratory demands of respiratory testing. Further work is required to define the optimal testing algorithm.