482: ROLE OF PNEUMONIA PCR PANEL ON METHICILLIN-RESISTANT STAPHYLOCOCCUS AUREUS COVERAGE IN A TRAUMA ICU

Abstract

Introduction: Pneumonia polymerase chain reaction (PCR) tests are a novel tool used to identify bacterial pathogens within hour(s) after sample collection versus up to 72 hours from a bronchoalveolar lavage (BAL). Early experience revealed a strong correlation between Staphylococcus aureus isolation and detection of resistance, and the trauma intensive care unit (TICU) adopted a protocol in September 2021 to escalate or de-escalate Methicillin-resistant Staphylococcus aureus (MRSA) coverage based on results. Methods: BioFire FilmArray pneumonia PCR panel assay results were evaluated in comparison to BAL culture results between September 2021 to March 2022 for patients admitted to the TICU of a large, academic, level 1 trauma center. Patients were categorized as early (< 5 days) or late (≥5 days) pneumonia in an ICU setting. The PCR panel assay was determined to correlate with the BAL if at least one organism identified in the PCR grew in the BAL. For patients treated for pneumonia, antibiotic regimens, antibiotic days, recurrence and clinical cure were evaluated and Fisher’s exact and Chi-square tests used for statistical analysis. Results: Pneumonia was suspected and PCR and BAL were obtained in 153 instances. Of those, 96 PCR panel assays detected bacteria (62.7%) with false negatives in 10/153 (6.5%) instances. Positive PCR panels correlated with BAL results in 52/96 (54.2%) of cases, with a positive predictive value (PPV) of 46.0%. 32 PCR assays detected S. aureus and 12 were detected in the corresponding BAL (PPV 37.4%). The presence or absence of methicillin-resistance correlated with cultures in all but one instance. Clinical outcomes were similar between groups for acute kidney injury, antibiotics days, MRSA days of coverage, recurrent pneumonia, and clinical cure. Conclusions: Pneumonia PCR panel assay demonstrated a 46% PPV with BAL culture results. The PCR assay demonstrates high sensitivity for S. aureus confirmed by BAL cultures and opportunities for de-escalation of MRSA coverage were identified in a majority of patients. Future educational efforts towards early antimicrobial streamlining may impact MRSA antibiotic duration, and larger sample sizes may reveal more correlation patterns.