Sustained impact of a rapid microarray-based assay with antimicrobial stewardship interventions on optimizing therapy in patients with Gram-positive bacteraemia.

Abstract

To compare antibiotic optimization and outcomes of patients before implementation of the Verigene Gram-Positive Blood Culture (Verigene BC-GP) nucleic acid microarray assay to after implementation with antimicrobial stewardship (AS) interventions and after discontinuation of AS interventions. A retrospective pre-post-post quasi-experimental study was conducted to compare the three periods. AS interventions consisted of real-time guidance to clinicians on antibiotic selection. The primary outcome was median time from Gram stain to optimal therapy. Secondary outcomes included median time to effective therapy, median duration of therapy for contaminant organisms, median length of stay after blood cultures were collected, and all-cause in-hospital mortality. Out of a total of 923 patients, 390 (125 baseline, 134 intervention, 131 post-intervention) who were not on optimal therapy at the time of Gram stain or had contaminated blood cultures were assessed. Compared with baseline, only the median time to optimal therapy for MSSA bacteraemia was reduced in both the intervention and post-intervention periods (17 versus 17 versus 50 h; P < 0.001), respectively. In an analysis adjusted for baseline differences among the groups using quantile regression models, use of the Verigene BC-GP assay in both periods significantly reduced time to optimal therapy by 14-22 h in patients who would achieve optimal therapy at ≥ 26 h without the assay. There were no differences in in-hospital mortality or hospital length of stay between study periods. A real-time AS intervention implemented alongside introduction of the Verigene BC-GP assay led to improvements in antibiotic therapy for patients with bacteraemia due to Gram-positive cocci, even after the AS intervention was discontinued.