Identification and Antibiotic Susceptibility Patterns of Clinical Blood Culture Isolates Not Identified by a Rapid Microarray Diagnostic System

The use of molecular-based diagnostic testing, such as the Luminex Verigene system, to rapidly identify the most common bacterial isolates from blood cultures is an important tool that reduces the duration of inappropriate antibiotics and decreases mortality. However, 5 to 15% of microorganisms recovered from blood culture are unable to be identified by the Verigene Gram-negative (BC-GN) or Gram-positive (BC-GP) assays. In this retrospective, observational study, we evaluate the identities and antimicrobial susceptibility patterns of 229 isolates that were not identified by either the Verigene BC-GN or BC-GP assay. The results presented here suggest that important, clinically relevant information about antimicrobial susceptibility patterns can still be inferred even when isolates are not identified by Verigene. We also examined changes in antibiotic use for patients with “unidentified” Verigene results at our institution and found that this subgroup represents an opportunity to optimize empirical antibiotic therapy. IMPORTANCE Rapid diagnostic testing to identify bloodstream pathogens has arisen as an important tool both to ensure adequate antimicrobial therapy is given early and to aid in antimicrobial stewardship by allowing for more rapid deescalation of inappropriate antimicrobial therapy. However, there is a paucity of data regarding the significance of isolates that are not able to be identified by rapid diagnostic testing. In this study, we report the identification to the species level and antimicrobial susceptibilities among isolates that were not identified by one such rapid diagnostic platform, the Verigene system. This study provides important insight into how a strong understanding of the strengths and limitations of a given rapid diagnostic platform, coupled with insight into local antibiotic susceptibility patterns, can allow for more nuanced and thoughtful empirical antibiotic selection.