Xpert Xpress Flu/RSV: Validation and Impact Evaluation at a large UK Hospital Trust.

Morris, Thomas Christopher; William Bird, Paul; Horvath-Papp, Eva; Kaur Dhillon, Jasminder; May, Shoshanna; Tang, Julian W

BACKGROUND Despite vaccination programmes and antivirals, influenza remains a prominent cause of morbidity and mortality. The Xpert Xpress Flu/RSV test is a leading influenza point-of-care test, but its evaluation has been limited to nasopharyngeal samples. In addition, the clinical impacts of Xpress Flu/RSV have not yet been quantified. METHODS We evaluated the performance of Xpress Flu/RSV at three locations in a UK Hospital Trust against an existing laboratory assay. Multiple upper respiratory tract sample types were included. In addition, we calculated time saved by Xpert, and the associations between Xpert use and rates of early patient isolation and antiviral prescription as recorded at the time of the laboratory result being telephoned out. RESULTS 642 patients were included in the diagnostic performance analysis. There were 177 laboratory-confirmed cases of influenza A, 7 influenza B and 86 RSV. For influenza A, sensitivity and specificity were 96.6% (95% confidence interval 92.8% - 98.8%) and 98.1% (96.4% - 99.1%) respectively. This was sustained across all locations and sample types. The negative predictive value was 98.7% (97.2% - 99.4%). The median amount of time saved was 27.1 hours. Xpert use was associated with six-fold higher rates of isolation and three-fold higher rates of antiviral prescribing by the time the laboratory result was available. Sensitivity for RSV was lower at 86.0% (95% CI: 76.9% - 92.6%). CONCLUSION Xpert Xpress Flu/RSV reliably detects influenza A infection and has significant clinical impacts. Cartridge optimisation is required to enable accurate multiplexing, including from a range of sample types. This article is protected by copyright. All rights reserved.