Retrospective analysis of pathogens for guided creation of an EMPIRIc antibiotic prEscribing pathway (EMPIRE)

Abstract

Abstract The objective of this study was to evaluate the susceptibilities of pathogens isolated from cultures within the first 7 days of admission to the burn center and in the absence of healthcare-associated infection risk factors (HAIRF) to determine if current empiric antibiotics can be narrowed for refinement of an empiric antibiotic prescribing pathway according to the suspected source. A 3-year sample of patients and cultures was utilized in hopes of obtaining at least 30 isolates of the most common pathogens and their respective susceptibilities. A total of 268 clinically relevant (e.g., deemed infectious, versus colonization) pathogens were included in the final sample with sources including wounds, respiratory, blood, urine, and bone. Of the 268 pathogens included, 45% were Gram-negative and 69% of all pathogens were isolated from wound cultures. The existing empiric pathway, vancomycin plus cefepime, covered 98% and 84% of all Gram-positive and Gram-negative pathogens, respectively. In patients without HAIRF, coverage rose to 98% and 90%, respectively. Initial use of vancomycin and cefepime remains adequate for pathogens isolated within 1 week of admission in patients without HAIRF. For pneumonias, a narrower spectrum beta-lactam would not sufficiently cover respiratory pathogens isolated within the first week of admission. Regarding early wound infections, difficult-to-treat pathogens remain as a rare isolate of wound cultures within 1 week of admission.