Variability of antibiograms: how often do changes in the antimicrobial susceptibility pattern occur in isolates from one patient?

Abstract

Objectives: Many laboratories do not perform antimicrobial susceptibility testing (AST) for all consecutive isolates of the same species from one patient. The objective of this study was to assess how often changes in antimicrobial susceptibility patterns (cASP) occur in such isolates.

Methods: AST was performed for all isolates of defined species obtained from clinical routine (2015-2018) without restrictions for consecutive or sequential isolates of one patient. Occurrence of cASP and time between the first sampling date and first cASP were determined by combining antibiograms from all specimens and after stratification into species and specimens.

Results: A total of 35 473 AST results were included (range 2-71 per case). Combining pathogens and specimens, 1991 cASP occurred in 1269/8502 (14.9%) of all cases after a median time until cASP of 5 days (range 0-364). Of these, 628/1991 (31.5%) occurred on the day of first sampling (predominantly due to phenotypic variants in the same specimen). Excluding isolates with differing AST pattern on the first day of sampling, the median time until cASP was 12 days (range 1-364). Stratification into species and specimen revealed a large variance of the median time until cASP (e.g. in Escherichia coli: 5 days; range 1-48 for blood cultures or 16 days; 1-364 for urine).

Discussion: Using routine microbiological data in a large tertiary hospital, cASP occurs occasionally. The time to perform subsequent AST to detect cASP depends on the species and type of specimen. Other studies are needed to evaluate whether ideal time intervals applicable beyond local settings can be defined.

Keywords: Antibacterial agents; Antimicrobial susceptibility testing; Costs; Resistance; Validation.