A Tertiary Care Center's Experience with Novel Molecular Meningitis/Encephalitis Diagnostics and Implementation with Antimicrobial Stewardship.

Background: Novel molecular techniques, such as the Biofire FilmArray Meningitis/Encephalitis (ME) panel, are increasingly used to improve pathogen detection and time to detection (TtD). The Brooke Army Medical Center antibiotic stewardship program evaluated the impact of the ME panel on empiric antimicrobial usage. Methods: Negative ME panels were analyzed for days of therapy (DOT). The ME panel became available at Brooke Army Medical Center on January 1, 2016 and a retrospective chart review was performed on all hospitalized patients tested by ME panel through April 30, 2016. Demographic data, cerebral spinal fluid (CSF) leukocyte count, immunocompromised status, and intensive care unit admission status were collected. TtD by ME panel and CSF culture were compared and DOT for common antimicrobials were quantified. Positive ME panels were analyzed for same demographic data, diagnoses, and microbiologic workup including CSF cultures and send out polymerase chain reactions. Results: Of the 77 ME panels performed during the study period, 54 (70%) were conducted on inpatients and included in the analysis. The majority of patients were males (n = 29, 54%) and the median age was 24 yr (interquartile range [IQR] 45; range 1 d to 83 yr). A total of eight (15%) patients were immunocompromised and 17 (31%) required intensive care unit level of care. The median TtD with the ME panel and CSF culture was 2.75 (IQR 2.16, 3.64) and 68.5 (IQR 63.87, 78.37) h, respectively. For negative ME panels, the overall median DOT for antimicrobials was 3 (IQR 1.5, 4.0) d, whereas the median DOT for individual agents was 2 (IQR 1.0, 4.0) d for vancomycin (n = 15), 1.5 (IQR 1.0, 2.25) d for ceftriaxone (n = 16), 3 (IQR 3.0, 4.0) d for ampicillin (n = 15), 3.5 (IQR 2.75, 4.0) d for gentamicin (n = 8), 3.5 (IQR 2.25, 4.0) d for cefotaxime (n = 6), and 5 (IQR 3.0, 5.5) d for acyclovir (n = 7); the median CSF leukocyte is of 2 cells/mm(3) (IQR 1.0, 7.5). DOT excluded cases of positive ME panels: human herpes virus-6 (n = 2), herpes simplex virus-2 (n = 3), enterovirus (n = 1), and Streptococcus pneumoniae (n = 1). Of these, there were two discordance diagnoses between ME panel and convention microbiologic methods. S. pneumonia was detected on the ME panel and not on the CSF culture. One bone marrow transplant recipient had symptoms of encephalitis caused by human herpes virus-6 detected only by the ME panel, the send out human herpes virus-6 polymerase chain reaction was negative. Conclusion: The ME panel appears to improve diagnostic yield in our facility, and there is potential for improvement in decreasing empiric antimicrobial usage, particularly in patients with a negative ME panel and absence of CSF pleocytosis. This demonstrates the need for antibiotic stewardship program involvement to assist in implementation of rapid diagnostic tests through methods such as education, clinical guidelines, and prospective audit and feedback to improve meningitis and encephalitis management.