Background: Impact assessment of multiplex point-of care tests (POCT) developed for the diagnosis of upper respiratory tract infection (RTI) on the frequency and duration of antimicrobial therapies in routine practice.
Methods: In this prospective multicentric observational study, data were collected on patients who underwent POCT on RTI admission in 10 participating centers. POCT were realized on BIOFIRE® FILMARRAY® RP 2 plus (bioMérieux) or ePlex® Respiratory Pathogen Panel (GenMark). For each patient, demographic data, admission diagnosis, potential antimicrobial treatments and the POCT results were collected. Dates and hours of admission, hospitalization, respiratory sampling, POCT results availability and, if used, antimicrobial treatment were also collected. To assess POCT impacts, comparison between test results and antimicrobial use was conducted between patients with POCT result available before and after antimicrobial prescription.
Results: Between November 2017 and December 2018 we enrolled 649 patients with a mean age of 56 yo [0;99]. Of these, 88% (n=558) were hospitalized: 58% in conventional wards and 42% in ICU. The most frequent admission causes were Influenza-like illness syndrome (26%; n=131) and pneumonia (19%; n=98). After exclusion of patients previously treated before their admission, 52% (n=265) and 7% (n=41) of patients received antibiotics and antivirals respectively. Among them, only 30% (n=82) of antibiotic treatments were prescribed upon test result availability, with a median time to antibiotics of 7.4 hours (IQR: 2.1;22.1) and a median time from consultation to result availability of 18.9 hours (IQR: 8.7;37.6). The positivity rates of POCT were 49% (IC95: 37-60%) versus 53% (IC95: 46%-60%, p=0.49) in patients with prescribed antibiotics before and after POCT results respectively. Moreover, mean durations of antibiotics were 9.9 (CI95: 8.6-11.1) days versus 10.4 (CI95: 9.0-11.7; p=0.58) when the POCT was positive versus negative respectively.
Conclusions: No association between antibiotic introduction nor antibiotic duration and POCT results were observed. Most (70%) antimicrobial treatments were introduced before POCT results. The potential lack of impact of POCT on antimicrobial therapies use could be explained by: (i) a long mean time to result incompatible with the necessities of emergency care, and (ii) limited physician reliance on POCT results.