Impact of rapid diagnostics with antimicrobial stewardship support for children with positive blood cultures: A quasi-experimental study with time trend analysis

• Published in: Infection control and hospital epidemiology 2020-08, Vol.41 (8), p.883-890
• Main Authors: Tribble, Alison C, Gerber, Jeffrey S, Bilker, Warren B, Lautenbach, Ebbing
• Format: Article
• Language: English
• Subjects: Antibiotics; Antimicrobial agents; Blood; Contaminants; Disease prevention; Gram-positive bacteria; Infectious diseases; Intensive care; Intervention; Length of stay; Medical diagnosis; Mortality; Nursing; Patients; Pediatrics; Pharmacists; Quasi-experimental methods; Staphylococcus infections; Time series; Trends; Variables
• ISSN: 0899-823X; 1559-6834
• DOI: 10.1017/ice.2020.191

Evaluate the clinical impact of the implementation of VERIGENE gram-positive blood culture testing (BC-GP) coupled with antimicrobial stewardship result notification for children with positive blood cultures. Quasi-experimental study. Quaternary children's hospital. Hospitalized children aged 0-21 years with positive blood culture events 1 year before and 1 year after implementation of BC-GP testing. The primary outcome was time to optimal antibiotic therapy for positive blood cultures, defined as receiving definitive therapy without unnecessary antibiotics (pathogens) or no antibiotics (contaminants). Secondary outcomes were time to effective therapy, time to definitive therapy, and time to stopping vancomycin, length of stay, and 30-day mortality. Time-to-therapy outcomes before and after the intervention were compared using Cox regression models and interrupted time series analyses, adjusting for patient characteristics and trends over time. Gram-negative events were included as a nonequivalent dependent variable. We included 264 preintervention events (191 gram-positive, 73 gram-negative) and 257 postintervention events (168 gram-positive, 89 gram-negative). The median age was 2.9 years (interquartile range, 0.3-10.1), and 418 pediatric patients (80.2%) had ≥1 complex chronic condition. For gram-positive isolates, implementation of BC-GP testing was associated with an immediate reduction in time to optimal therapy and time to stopping vancomycin for both analyses. BC-GP testing was associated with decreased time to definitive therapy in interrupted time series analysis but not Cox modeling. No such changes were observed for gram-negative isolates. No changes in time to effective therapy, length of stay, or mortality were associated with BC-GP. The implementation of BC-GP testing coupled with antimicrobial stewardship result notification was associated with decreased time to optimal therapy and time to stopping vancomycin for hospitalized children with gram-positive blood culture isolates.