Performance of BIOFIRE FILMARRAY pneumonia panel in suspected pneumonia: insights from a real-world study

Abstract

Pneumonia may be caused by a diverse group of microorganisms; however, a microbiologic diagnosis is not universally made. Molecular tests such as the BIOFIRE FILMARRAY Pneumonia Panel (BF-PP) offer the possibility of rapid identification of potential pneumonia pathogens. This retrospective analysis was conducted as a sub-study of a recently published randomized controlled trial. Specimens from patients with suspected pneumonia were tested by the BF-PP, with results compared to cultures. The overall BF-PP positivity rate was 44.8% (252/563), and was higher [64.6% (164/254)] in specimens from patients ultimately diagnosed with pneumonia. Discrepancies between the BF-PP and cultures were most common for Streptococcus pneumoniae, Haemophilus influenzae, Moraxella catarrhalis, and Streptococcus agalactiae. Positive percent agreement (PPA) and negative percent agreement (NPA) between the BF-PP and culture were 91.4% (CI 86.3%-94.9%) and 83.6% (CI 79.5%-87.2%), respectively. For most bacteria, a correlation between high genomic abundance (genomic copies/mL) reported from the BF-PP and culture growth abundance was found. In specimens from patients ultimately diagnosed with pneumonia, clinical consistency-assessed based on whether BF-PP results aligned with associated culture results-was 79.9% and was highest for bronchoalveolar lavage fluid specimens. The BF-PP has reliable analytical performance and offers the potential advantage over conventional cultures of providing higher detection rates and more rapid results. Interestingly, despite the use of this advanced molecular diagnostic tool, one-third of pneumonia cases remained without a microbiologic etiology.IMPORTANCEThis study evaluates the BIOFIRE FILMARRAY Pneumonia Panel by comparing its performance to conventional cultures in a real-world patient population (including patients ultimately diagnosed as not having pneumonia) using different types of specimens. Findings show a higher detection rate of microorganisms with the panel compared to cultures, suggesting that this test could aid in tailoring treatments for pneumonia. However, challenges remain and require further study, including distinguishing true pathogens from colonizing microorganisms and determinig why one-third of patients diagnosed with pneumonia still lacked a microbiologic etiology.

Keywords: BIOFIRE; analytical performance; concordance; negative percent agreement; pneumonia; positive percent agreement.