Performance evaluation of a commercial multiplex pathogen panel for detection of bacteria in sputum specimens from non-ICU patients with suspected lower respiratory tract infection

Rapid diagnostic testing can improve pathogen detection and lead to targeted antibiotics. The BioFire FilmArray Pneumonia Panel (BFPP) is a multiplex PCR that has displayed strong concordance with traditional microbiologic techniques. However, most existing literature focuses on deep respiratory specimens, and there is sparse literature on performance in sputum specimens. This retrospective, single-center study included adult patients between 1 September 2022 and 31 August 2024 who had collection of a BFPP with standard of care (SOC) culture from a sputum specimen on a non-intensive care unit (ICU) floor or in the emergency department if admitted to a non-ICU floor. Out of 189 BFPPs performed on 189 sputum specimens, a total of 141 bacterial targets were detected. Between the BFPP and SOC culture, the overall positive percent agreement and negative percent agreement (NPA) were 96.3% and 54.9%, respectively. The positive predictive value (PPV) was 26.3% while the negative predictive value was 98.9%. Patients with greater than 24 h of antibiotic exposure prior to BFPP collection had a lower PPV compared to patients with less than 24 h or no exposure (13.6% vs 29.6% vs 30.4%). The lowest concordance was observed for Haemophilus influenzae (15.4%), Moraxella catarrhalis (18.2%), Streptococcus pneumoniae (19%), and Staphylococcus aureus (22.7%), several of which are fastidious in culture. BFPP showed a high NPA, with all bacterial targets having an NPA greater than 90%, except H. influenzaeinfluenzae(82%). Based on these data, a negative BFPP in sputum specimens could help to rule out a bacterial pneumonia, but the benefit of a positive test remains unclear. IMPORTANCE This study evaluates the BioFire FilmArray Pneumonia Panel (BFPP) by comparing its performance to standard of care cultures exclusively in sputum specimens from non-intensive care unit patients with suspected lower respiratory tract infection. Findings show an overall high positive percent agreement and negative predictive value but a low negative percent agreement and positive predictive value, suggesting that a negative test in sputum specimens could be beneficial when attempting to rule out a bacterial infection, but the benefit of a positive test remains unclear, particularly if common airway colonizing bacteria are detected and at low semi-quantitative thresholds. Clinical symptoms should guide test interpretation in patients with positive BFPP results but negative culture growth.