643. Comparison of Multiplex Polymerase Chain Reaction (PCR) and Routine Culture for the Detection of Respiratory Pathogens in Pneumonia Patients

BACKGROUND: The identification of causative pathogens in pneumonia can be challenging, and conventional culture methods can take up to 72 hours. However, rapid microbiologic tests identify organisms within hours. The Biofire®Filmarray ((bioMérieux, North Carolina) Pneumonia Panel was recently approved by the FDA. The multiplex PCR system identifies 33 targets from sputum and bronchoalveolar (BAL) samples, which include 18 bacteria, 8 viruses, and 7 antibiotic resistance genes. The purpose is to compare the panel to routine culture methods for the detection of respiratory pathogens in patients with pneumonia in a 794-bed teaching hospital in northwest Ohio. METHODS: We retrospectively screened all hospitalized intensive care unit patients who met clinical and radiological criteria of pneumonia using electronic medical records between November 2018 and February 2019. Adult patients who had respiratory cultures collected within 7 days were included. Repeat specimens were excluded. Routine cultures were performed using the laboratory’s standard procedure, and Pneumonia Panel testing was performed according to manufacturer instructions. RESULTS: Fifty-nine respiratory or 13 BAL and 46 sputum specimens were evaluated. There was no discrepancy between culture and PCR in 63% (37/59) samples. One (8%) BAL and 10 (22%) sputum specimens had additional pathogens detected by PCR. There was a discrepancy between culture and PCR in four (31%) BAL and seven (15%) sputum samples. The largest discrepancy was noted amongst Serratia marcescens (4/59 or 7%) and Haemophilus influenzae (6/59 or 10%) species. Only one sputum culture had Legionella detected by PCR. Additionally, 17 specimens had a virus detected either alone or with another bacterial pathogen by PCR. For the resistance genes, KPC was detected by PCR but not by Modified Carbapenem Inactivation Method (mCIM) test. The mecA gene was detected in six of seven (86%) of methicillin-resistant Staphylococcus aureus (MRSA) isolates. CTX-M was detected in Serratia and Klebsiella pneunomiae in two samples; however, the organisms were not isolated in culture. CONCLUSION: The Pneumonia Panel can identify additional bacteria that did not grow in culture. This panel can rapidly identify pathogens and potentially reduce unnecessary antibiotic use. DISCLOSURES: All authors: No reported disclosures