57

Abstract

Introduction: Intensive Care Unit (ICU) providers are challenged with the emergence of multi-drug resistant pathogens such as Pseudomonas aeruginosa (PsA). ICU specific antibiograms are commonly used to track local resistance patterns and to help guide empiric antimicrobial therapy. Antibiograms are developed using standards and antibiotic minimum inhibitory concentration (MIC) breakpoints set by the Clinical and Laboratory Standards Institute (CLSI). Recently, some of the CLSI susceptibility breakpoints for aminoglycosides, certain beta-lactams and fluoroquinolones have been called into question secondary to pharmacokinetic (PK) and pharmacodynamics (PD) data. The CLSI susceptibility breakpoints may be set too high, which could lead to sub-optimal therapy. Hypothesis: If more stringent PK/PD susceptibility breakpoints are used for construction of antibiograms more resistance will be observed than if CLSI susceptibility breakpoints are used for PsA. Methods: A retrospective study was conducted on PsA isolates collected from adults 18 years and older in the medical ICU (MICU) and surgical ICU (SICU) from 01/08 to 12/10. Only the first isolate collected from blood or respiratory sources were considered. Unit-specific antibiograms were developed using the 2011 CLSI susceptibility breakpoints and compared to antibiograms developed with more stringent PK/PD breakpoints. Results: A total of 211 PsA isolates were included. Tobramycin had the highest susceptibility in the two units regardless if CLSI (99% in MICU, 100% in SICU) or PK/PD (88% in MICU, 98% in SICU) breakpoints were used. In the PK/PD group ceftazidime had the highest susceptibility amongst the beta-lactams in both SICU (79%) and MICU (82%). Piperacillin/tazobactam (P/T) had the highest susceptibility amongst the beta-lactams in the CLSI groups; however, the susceptibility decreased when the more stringent PK/PD breakpoint was used in both MICU (90% vs. 66%, p=0.001) and SICU (92% vs. 65%, p=0.001). A 28% (p=0.001) and 35% (p=0.001) absolute decrease in susceptibility was observed with ciprofloxacin between the CLSI and PK/PD groups in the MICU and SICU, respectively. Conclusions: The observed resistance rates were higher for P/T, ciprofloxacin and the aminoglycosides when PK/PD derived susceptibility breakpoints were used for development of antibiograms instead of the 2011 CLSI breakpoints.