Challenges in establishing epidemiological cut-off values for the Burkholderia cepacia complex

The Burkholderia cepacia complex (BCC) is comprised of 24 species of gram-negative bacteria that cause opportunistic infections. While antimicrobial susceptibility testing (AST) has historically been used to guide treatment for BCC infections, recent work highlighting problems with AST for these organisms led the Clinical and Laboratory Standards Institute (CLSI) to remove disk diffusion and minimal inhibitory concentration (MIC) breakpoints for BCC from its M100 standards document. Epidemiological cut-off values (ECVs) may be helpful to clinicians in the absence of breakpoints, as they may be used to determine whether an isolate has a wild-type or non-wild-type phenotype. Here, we present an analysis of BCC ECVs for ceftazidime (CAZ), levofloxacin (LVX), meropenem (MEM), minocycline (MIN), and trimethoprim-sulfamethoxazole (TMP-SMX). ECVs were calculated using MIC data from 3 previous studies and 3 independent laboratories for 1,896 BCC isolates. ECVs were 16 μg/mL for CAZ, 8 μg/mL for LVX, 16 μg/mL for MEM, and 8 μg/mL for MIN. The ECV for TMP-SMX varied depending on the analysis from 2, 8, and 16 μg/mL and therefore could not be reliably established. Challenges with establishing ECVs for BCC include limitations with the pooled MIC data set, broad MIC distributions, and high ECVs that are above the obsolete susceptible MIC breakpoints. These challenges limit the clinical utility of ECVs for these organisms and support the removal of ECVs from the CLSI M100 standards document.

The Burkholderia cepacia complex is a group of bacterial species that cause difficult-to-treat opportunistic infections. Recently, clinical breakpoints, which are used to determine whether organisms are susceptible to certain antimicrobials, were removed from Clinical and Laboratory Standards Institute (CLSI) standards for these organisms due to problems with antimicrobial susceptibility testing performance. Clinicians are now faced with the challenge of how to treat these complex infections without clinical breakpoints. Here, we determine epidemiological cut-off values (ECVs) for relevant antimicrobials for the B. cepacia complex. While we established ECVs for ceftazidime, levofloxacin, meropenem, and minocycline, we were unable to determine a reliable ECV for trimethoprim-sulfamethoxazole. Additionally, we encountered significant challenges in our analyses, including data limitations for these organisms and high ECVs that are not clinically useful. These challenges limit the practical use of these ECVs in helping guide treatment and support the removal of ECVs from the CLSI M100 standards document.