648 - BioFire^® FilmArray^® Pneumonia plus Panel Performance Evaluation: A Multicenter, International Collaborative Study

Background:

Background: Classical methods to identify causes of community acquired, healthcare and ventilator associated pneumonia can be insensitive and slow, leading to unnecessary or inappropriate antimicrobial therapy. The BioFire^® FilmArray^® Pneumonia plus Panel (PNplus) detects 15 bacteria (in semi-quantitative log bin values from 10^4 to > 10^7), 7 antibiotic resistance markers (mecA/C/MREJ, CTX-M, KPC, VIM, IMP, NDM, OXA-48 like), 3 atypical bacteria (AB), and 8 viral classes directly from bronchoalveolar lavage (BAL)-like and sputum-like specimens (including endotracheal aspirates) in about 1 hr.  This study compared PNplus results to standard of care testing (SOC).

Methods:

Methods: 2476 samples (1234 BAL-like; 1242 sputum-like) were tested at 52 laboratories from 13 European countries and Israel by PNplus and SOC. SOC varied by site and physician prescription.  Pathogen detection rates were compared. PNplus bin values and SOC descriptive or numerical quantities were evaluated for 1297 bacterial detections.

Results:

Results: 13 samples (0.5%) gave invalid PNplus results. 3278 bacteria in PNplus were detected by PNplus and/or SOC. SOC detected 1878 bacteria (57.1%) compared to 3128 bacteria (95.8%) for PNplus (p=< 0.0001).  SOC detected 73 AB (70.9%) and 134 viruses (21.1%), PNplus detected 93 AB (90.3%) and 618 viruses (97.9%) (p=< 0.0001). Mean number of analytes/sample detected by PNplus and SOC were 1.99 and 1.44, respectively.  PNplus bin values were less than SOC, equal to SOC or greater than SOC in 5.9%, 25.4% and 69.6% of results, respectively. PNplus values were on average > 1 log than SOC values (58.5% 1-2 logs; 11.0% 3-4 logs). PNplus identified 98.2% of MRSA and SOC 55.6%.  All gram-negative resistance markers were detected at least once. PNplus and SOC results were fully concordant (positive or negative) or partially concordant for 49.1% and 26.4% of samples, respectively.

Conclusion:

Conclusions: PNplus detected significantly more potential pathogens than SOC. Lack of routine SOC viral testing was a missed opportunity to define the cause of pneumonia. Semi-quantification may assist in understanding the significance of the pathogens detected. Pathogen and resistance marker detection in about 1 hr could dramatically impact antimicrobial use and enhance patient outcomes.