Clinical Study
Clinical and economic impact of antimicrobial stewardship interventions with the FilmArray blood culture identification panel

https://doi.org/10.1016/j.diagmicrobio.2015.10.023Get rights and content

Highlights

  • We report clinical use of the FilmArray Blood Culture Identification (BCID) panel.

  • Antibiotic use was improved in comparison to matched historical controls.

  • Patients with contaminated blood cultures were discharged earlier.

  • The benefit of the BCID may depend on baseline antimicrobial stewardship activity.

Abstract

The purpose of this study was to evaluate the impact of the FilmArray Blood Culture Identification (BCID) Panel on the management of patients with blood cultures growing gram positive cocci and Candida. We retrospectively compared clinical and economic outcomes between patients during the BCID testing period and a matched historical control group before BCID testing was introduced. A total of 84 BCID patients were matched to 252 historical controls. BCID identification of coagulase negative staphylococci contaminants resulted in shorter post-culture length of stay (P < 0.008) and saved roughly $30,000 per 100 patients tested. The BCID led to shorter duration of empirical vancomycin for patients with contaminated blood cultures (P = 0.005) and methicillin-susceptible Staphylococcus aureus bacteremia (P < 0.001). Patients with vancomycin-resistant enterococcal bacteremia received active therapy earlier than historical controls (P = 0.047). The BCID, coupled with antimicrobial stewardship intervention, was a cost effective tool to improve patient care.

Introduction

Antimicrobial stewardship programs (ASP) optimize infection-related patient outcomes while minimizing unintended consequences of treating infections (i.e., emergence of antimicrobial resistance, adverse drug reactions) (Dellit et al., 2007). The initiation of appropriate antimicrobial therapy and antimicrobial de-escalation both hinge on timely provision of pathogen identification and susceptibility results from the clinical microbiology laboratory. Antimicrobial therapy cannot be streamlined until a target is identified.

Standard blood cultures provide critical information, but the process is not ideal. Complete organism identification and susceptibility testing generally takes 2–5 days after sample collection, necessitating several days of empirical therapy. There are several problems with the current paradigm of empirical therapy. Empirical therapy can be inadequate or suboptimal, risking poor patient outcomes (Kumar et al., 2006). On the other hand, empirical therapy contributes to the overuse of broad-spectrum agents and is frequently provided for patients without infections. Blood culture contamination contributes to empirical antimicrobial exposure, prolonged hospitalizations, and increased hospital expenditures (Alahmadi et al., 2011).

The FilmArray Blood Culture Identification (BCID) Panel (BioFire Diagnostics, Salt Lake City, UT) is a multiplexed polymerase chain reaction (PCR)-based diagnostic test cleared for use with positive blood cultures (BioFire Diagnostics Inc., 2013). The BCID detects 19 bacteria, five Candida spp. and three antimicrobial resistance determinants (mecA, vanA/B, blaKPC), in roughly 1 h from the time of culture positivity (Table S1). We were interested in using the BCID test to rapidly 1) identify gram positive cocci in clusters as coagulase-negative staphylococci (CoNS), methicillin-sensitive Staphylococcus aureus (MSSA), or methicillin-resistant S. aureus (MRSA), 2) identify enterococci as vancomycin-resistant (VRE) or vancomycin-sensitive (VSE) and 3) identify azole-resistant Candida spp. based upon species specific resistance patterns.

Previous studies have established the diagnostic accuracy of the BCID in comparison to conventional laboratory methods (Altun et al., 2013, Bhatti et al., 2014, BioFire Diagnostics Inc., 2013, Blaschke et al., 2012, Rand and Delano, 2014, Ward et al., 2015). However, published studies have not described use of the BCID to guide clinical decision making through an ASP in real time. The purpose of this study was to evaluate the clinical and economic impact of BCID testing with ASP intervention for patients with gram positive bacteremia and candidemia.

Section snippets

Study design and setting

This was a single center, pre-post intervention quasi-experimental study conducted at University of Florida (UF) Health Shands Hospital, a 939-bed academic medical center in Gainesville, Florida. The intervention group was 84 adult patients with gram positive bacteremia and/or candidemia identified via the FilmArray BCID between August 1, 2013 and January 31, 2014. This group was compared with a matched historical control group with organism identification and susceptibility testing performed

Results

During the BCID period, 103 patients met study criteria and had specimens processed on the BCID. A total of 84 BCID patients were matched to 252 historical controls (10 streptococcal infections and 9 true CoNS bacteremias were not included in the matched analyses). Baseline characteristics were similar between groups (Table 1), including comparisons within organism subgroups. Roughly half of the positive cultures in the BCID group, 46/84 (55%), represented CoNS contamination. Fifty percent of

Discussion

Patients with suspected bloodstream infection are an ideal population to target with rapid diagnostic tests. Some patients receive inadequate empirical therapy, while some receive therapy in the absence of infection. Our ASP used rapid results from the FilmArray BCID Panel to optimize therapy across a broad adult population, including those with contaminated blood cultures as well as those with true gram positive and Candida bloodstream infections. Our results demonstrate the “real world”

Acknowledgements and disclosures

We thank Donald Sterner of UF Health Decision Support Services and the UF Health Shands Hospital Clinical Microbiology Laboratory team.

Study funded in part by BioFire Diagnostics, Salt Lake City, UT. The funding source did not participate in study design or manuscript preparation. K.K. has received honoraria for participation on the advisory board for Astellas Pharmaceuticals and The Medicines Company. K.R. has received honoraria for speaking at the Scientific Advisory Board Meeting for BioFire

References (22)

  • N. Dhiman et al.

    Evaluation of the BinaxNOW Staphylococcus aureus test for rapid identification of gram-positive cocci from VersaTREK blood culture bottles

    J Clin Microbiol

    (2013)
  • Cited by (92)

    • Blood culture identification (BCID) performance in polymicrobial bacteremia

      2023, Diagnostic Microbiology and Infectious Disease
    • Concordance of the filmarray blood culture identification panel 2 and classical microbiological methods in a bacteriemia diagnostic unit

      2022, Diagnostic Microbiology and Infectious Disease
      Citation Excerpt :

      This continuous timetable of the Microbiology Laboratory has demonstrated its usefulness in reducing the rates of ineffective empirical therapies. However, each center should evaluate the techniques to be developed according to its characteristics, highlighting the importance of assessing the local epidemiology of bacterial resistance [14-16]. The good concordance of the technique with the classical methods of our research corroborates the previously published data, both at the level of etiology of the process and detection of resistance mechanisms.

    • Initial Specimen Diversion Device® reduces blood culture contamination and vancomycin use in academic medical centre

      2022, Journal of Hospital Infection
      Citation Excerpt :

      As an additional QI measure by the antibiotic stewardship programme committee, a retrospective study evaluating the reduction in vancomycin DOT per 1000 patient-days from ED admissions for sepsis following the implementation of a NAAT (Verigene, Luminex Corporation, Northbrook, IL, USA) (for rapid identification of organisms) and later SP was completed. We found that the NAAT reduced vancomycin DOT by 18.4% over an eight-month period before the start of the ISDD project in the ED, as others have reported [45,46]. There was significantly, however, a further incremental 31.4% reduction in vancomycin DOT over the following eight-month period after SP had been introduced.

    View all citing articles on Scopus
    1

    Present Address: North Florida/South Georgia Veterans Health System, 1601 SW Archer Road, Gainesville, FL 32608, USA.

    View full text