[Performance evaluation of VITEK 2 system in meropenem susceptibility testing of clinical Pseudomonas aeruginosa...

Pseudomonas aeruginosa is an important opportunistic pathogen associated with various community-acquired or nosocomial infections. Multi-drug resistant P.aeruginosa strains increasingly cause epidemics and spread in various hospital wards and geographic regions. Carbapenems are among the most effective antimicrobials in the treatment of multi-drug resistant P.aeruginosa infections, and meropenem is the most successful among alternatives in initial therapy. Particularly in severe infections, inappropriate or inadequate initial antimicrobial therapy is independently associated with adverse clinical and economic outcomes. Availability of accurate and rapid susceptibility testing is a priority. Most of the automated microbiology systems can provide rapid results within 8 to 12 hours. In comparison to standard methods, problems in the antimicrobial susceptibility testing of particular microorganisms and antimicrobial agents have been reported for automated microbiology systems. Failures have been reported previously especially in the susceptibility testing of P.aeruginosa versus carbapenem. Most of these studies are designed according to the Food and Drug Administration (FDA, USA) performance analysis scheme (Class II Special Controls Guidance Document: Antimicrobial Susceptibility Test Systems) in a simplified form. However, there are many lacking issues in the design of most of these studies. Among these, insufficient sample size, use of inappropriate reference method, lack of reproducibility testing, and inadequate distribution of study isolates in interpretative categories are of notice. There are only few studies in the literature that evaluate the performance of automated systems in antimicrobial susceptibility testing of carbapenems in clinical P.aeruginosa isolates with a sufficient sample size (n ? 100). However, most of these studies still have one or more major deficiencies in the study design. Furthermore, none of these studies evaluate the performance of VITEK 2 system without a major deficiency in study design. Therefore, we aimed to evaluate the performance of VITEK 2 system (bioMérieux, France) in antimicrobial susceptibility testing of carbapenems in clinical P.aeruginosa isolates in a well-designed study. The study was conducted on nonrepetitive P.aeruginosa isolates (n= 142) of clinical origin. Isolates were selected from the isolate collections of Culture Collection Unit of the Medical Laboratories at Ondokuz Mayis University Hospital. The study collection was characterized with conventional tests and the VITEK 2 automated microbiology system. Broth microdilution method standardized by Clinical and Laboratory Standards Institute was used as the reference method. P.aeruginosa ATCC 27853 was used as the quality control strain in all experimental steps. Twofold dilutions of meropenem (AstraZeneca, USA) concentrations between 64 mg/L and 0.125 mg/L were tested. In compliance with FDA recommendations, minimum inhibitory concentrations of study isolates were shown to be on-scale and distributed within the range of five sequential dilutions in both methods. In reproducibility testing, 15 organisms were tested with VITEK 2 system in triplicate. Results of the reproducibility tests were evaluated in comparison to the test mode (the most frequent test result for the isolate) as a reference. Overall reproducibility was 100%. Essential and categorical agreements of the VITEK 2 system in comparison to the reference method were 83.8% and 96.5%, respectively. Very major and minor discrepancy rates were 1.4% and 2.8%, respectively. There was no major discrepancy. While the results of the essential agreement was not acceptable, categorical agreement was acceptable according to the FDA performance criteria. There was very good agreement between methods as shown by the kappa value (?= 0.938). In conclusion, VITEK 2 system exhibited acceptable performance in the meropenem susceptibility testing of clinical P.aeruginosa isolates. As pre-market approval may not guarantee proper validation, performance of the automated microbiology systems in antimicrobial susceptibility testing should at least be verified and the literature that reports performance evaluation results should be read critically before implementation for routine use in laboratory.