# P-816. Are We Selecting the Right Patients for FilmArray Blood Culture Testing? Insights from a Gram-Negative Bacteremia Cohort

## TL;DR

This study examines whether the right patients are being selected for a costly blood test that can guide antibiotic treatment in Gram-negative bacteremia.

## Contribution

The study identifies factors associated with antibiotic optimization opportunities and evaluates the appropriateness of current test usage.

## Key findings

- Opportunities for antibiotic optimization were found in 31.8% of patients and linked to higher mortality.
- Only 59.3% of these optimization cases were tested using the FilmArray panel.
- Variables like male sex and comorbidities were associated with optimization but did not influence test selection.

## Abstract

Gram-negative bacteremia (GNB) is associated with high mortality. The implementation of FilmArray blood culture panels (FAP) has been shown to reduce mortality in patients with sepsis. However, the high cost of these panels limits their routine use in all GNB cases. This study aimed to identify variables associated with opportunities for antibiotic optimization (OAO) using the FAP in GNB, and to evaluate the appropriateness of its current utilization.

Ambispective cohort study of patients with GNB hospitalized in general wards at an Argentine hospital. Blood samples were collected using BactAlert® blood culture bottles. Microorganism identification and antimicrobial susceptibility testing were performed using the VITEK® 2C automated system. The FilmArray® BioFire® BCID2 panel was applied selectively, based on clinical judgment by the treating physicians. Opportunities for antibiotic optimization (OAO) using the FAP were defined as the detection of microorganisms or resistance genes likely to require modification of empirical therapy. OAO criteria included the presence of carbapenemase genes, special pathogens (such as Listeria spp or Salmonella spp.) and, in community-acquired infections, Pseudomonas aeruginosa and CTX-M-type extended-spectrum beta-lactamase (ESBL) genes. For statistical analysis, Chi2 and Mann–Whitney U tests were used, and variables with p< 0.05 were included in multivariate logistic regression models.

85 patients were included. FAP was used in 38.8% (n=33) of cases. Variables associated with FAP use are presented in Table 1. OAO were identified in 31.8% (n=27) of the cohort, and were associated with higher mortality (40.7% vs 17.2%, p=0.019). Only 59.3% of OAO cases were tested using the FAP. Variables associated with OAO are shown in Table 2.

In this cohort, FAP use was more frequent in severely ill patients, as indicated by higher SOFA scores and prior Intensive Care unit (UCI) admission. However, broader application may be warranted due to GNB high baseline mortality. Variables significantly associated with OAO—including male sex, presence of comorbidities, community-onset infections, ICU stay during the current hospitalization and prior hospital admissions—did not influence the selection of patients for FAP use.

All Authors: No reported disclosures