Rapid surveillance of multidrug resistance in ICU patients, by use of antibiotic-containing agar plates
Abstract number: P2069
Poulakou G., Plachouras D., Kontopidou F., Antoniadou A., Papadomichelakis E., Armaganidis A., Giamarellou H.
Objective: Adequate and timely administered antimicrobial treatment is the cornerstone of successful outcome in ICU infections. The aim of the study was to assess the efficacy of an in vitro method in the early screening of multidrug-resistant Gram-negative microorganisms colonising or infecting ICU patients, in view of guiding initial treatment.
Methods: In an 18-bed general ICU, surveillance cultures were performed bi-weekly for tracheal aspirates and weekly for urine and anal swabs. All specimens were directly inoculated on Mac-Conkey agar plates containing antibiotics at a concentration of 8 mg/L for Ceftazidime (CAZ), 1 mg/L for Ciprofloxacin (CIP), 4 mg/L for Imipenem (IMP) and 16/4 mg/L for Piperacilline/Tazobactam (PT). The method was evaluated for a six week period and was compared to standard microbiological procedures (disk diffusion method, CLSI 2007 breakpoints of resistance). For COL BSAC breakpoint for disk diffusion method was used. Tracheal aspirates were processed quantitatively. Discordant results were categorised as major error when a resistant isolate did not grow on the antibiotic containing agar, appearing to be sensitive, and as minor when the converse occurred.
Results: A total of 604 Gram-negative isolates were compared; 208 Acinetobacter baumannii, 138 Pseudomonas aeruginosa, 27 Stenotrophomonas maltophilia and 271 enterobacteriaceae (115 Klebsiella pneumoniae). Concordance of the two methods, minor and major errors of the tested antibiotic-containing agar plates are displayed in table 1. Specimens with high dense inocula did not affect the sensitivity of the method for all antibiotics except PT (p 0.006). The presence of multiple microorganisms in the specimen reduced significantly the sensitivity of CAZ- and PT-enriched plates (p 0.005 and 0.014 respectively). COL-enriched plates displayed unacceptably reduced concordance with standard method, probably reflecting the poor diffusion of the antibiotic in the agar.
In 12 enterobacteriaceae the final result of CAZ disk was reported discordant to that measured by diameter due to an ESBL phenotype, whereas this was also the case for IMP disk among 30 isolates of Pseudomonas aeruginosa and enterobacteriaceae with MBL phenotype.
Conclusions: The use of antibiotic-containing agar plates demonstrates satisfactory concordance with standard microbiological procedures and could prove useful in the early guidance of initial treatment of ICU patients with infections.
|Session name:||18th European Congress of Clinical Microbiology and Infectious Diseases|
|Back to top|