Abstract
BACKGROUND: Antimicrobial resistance (AMR), including multidrug (MDR) and extensively drug-resistant (XDR) bacteria, is an essential consideration in the prevention and management of hospital-acquired pneumonia (HAP) and ventilator-associated pneumonia (VAP). In the AMR era, the clinical utility of the BioFire FilmArray Pneumonia Panel Plus (BFPP) to diagnose HAP/VAP has not been thoroughly evaluated. METHODS: We enrolled adult hospitalized patients with HAP or VAP at Siriraj Hospital and Saraburi Hospital from July 2019-October 2021. Respiratory samples were collected for standard microbiological assays, antimicrobial susceptibility testing (AST), and the BFPP analysis. RESULTS: Of 40 subjects, 21 were men. The median duration of HAP/VAP diagnoses was 10.5 (5, 21.5) days, and 36 endotracheal aspirate and 4 sputum samples were collected. Standard cultures isolated 54 organisms-A. baumannii (37.0%), P. aeruginosa (29.6%), and S. maltophilia (16.7%). 68.6% of Gram Negatives showed an MDR or XDR profile. BFPP detected 77 bacterial targets-A. baumannii 32.5%, P. aeruginosa 26.3%, and K. pneumoniae 17.5%. Of 28 detected AMR gene targets, CTX-M (42.5%), OXA-48-like (25%), and NDM (14.3%) were the most common. Compared with standard testing, the BFPP had an overall sensitivity of 98% (88-100%), specificity of 81% (74-87%), positive predictive value of 60% (47-71%), negative predictive value of 99% (96-100%), and kappa (κ) coefficient of 0.64 (0.53-0.75). The concordance between phenotypic AST and detected AMR genes in Enterobacterales was 0.57. There was no concordance among A. baumannii, P. aeruginosa, and S. aureus. CONCLUSIONS: The BFPP has excellent diagnostic sensitivity to detect HAP/VAP etiology. The absence of S. maltophilia and discordance of AMR gene results limit the test performance.