[Spread of armA 16S rRNA Methylase Gene in a Tertiary Hospital and Errors in Aminoglycoside Susceptibility Results of Rapid Susceptibility Test Systems].

Aminoglycosides (AGs) are actively used in combination therapies against carbapenem resistant gram negative species in recent years. Spread of 16S rRNA methylases which can cause high-level resistance to AG antibiotics, limits this treatment choice. Although there are some studies showing that errors in determining AG susceptibility in automated systems may be related to the armA gene, one of the 16S rRNA methylase genes, the exact reason for these errors is not yet known. In our study, we aimed to investigate the relevance of 16S rRNA methylases to the discrepancies between VITEK 2.0 and disc diffusion test results for amikacin (AK) and gentamicin (GEN) susceptibility of Acinetobacter baumannii and Klebsiella pneumoniae isolates. All K.pneumoniae and A.baumannii isolates from 1st January-10th February 2018 were collected prospectively and included in the study. Additionally, two initial isolates from July 2017 (one K.pneumoniae and one A.baumannii isolate) for which first discrepant susceptibility results were determined, were also included. Amikacin and gentamicin susceptibility results of 37 isolates [A. baumannii (n= 20) and K.pneumoniae (n= 17)] were evaluated together with VITEK 2.0 system, disc diffusion and gold standard broth microdilution methods and minor error (mE), major error (ME) and very major error (VME) rates were calculated. The rmtB, rmtC and armA genes in isolates were investigated by polymerase chain reaction (PCR) and the relationship between the presence of 16S rRNA methylases and false susceptibility results were examined. In addition, disc diffusion test results were evaluated at the end of four, six, eight hours and one night incubation periods to examine the effect of the double zone phenotype observed in 13 of the study isolates on rapid susceptibility tests. All disc diffusion test results were found to be compatible with broth microdilution test results. When the VITEK 2.0 system and the broth microdilution test were compared, 10.3% and 12.1% VME and 8.1% and 5.4% mE were detected for AK and GEN susceptibility results, respectively. While rmtB and rmtC genes were not detected in the study isolates, armA gene was positive in eight (47.1%) of 17 K.pneumoniae isolates and in 15 (75%) of 20 A.baumannii isolates. All three VMEs in A.baumannii isolates were detected in AK susceptibility results. Two of those were armA gene positive and one was armA gene negative isolates. All four VMEs in K. pneumoniae isolates were detected in GEN susceptibility results only, and all of these isolates were armA gene positive. No direct correlation was found between the errors detected in the VITEK 2.0 system susceptibility results and the double zone phenotype. When the isolates were evaluated in the 4-16 hours incubation time interval, it was observed that resistant colonies could be detected after a minimum of six hours of incubation period in the inhibition zone surrounding the aminoglycoside discs. To the best of our knowledge this is the first report of armA producing A.baumannii from Turkey. The high rate of armA gene positivity detected in our isolates suggested that the prevalence of armA gene increased in our country or at least in our region, in recent years. In the AG susceptibility results of the VITEK 2.0 system, the rate of VME above the acceptance criterion has shown that the errors occurred were not directly related to armA gene positivity or double zone phenotype. Finally, our study results indicated that AG susceptibility results should be evaluated minimum six hours later of incubation while implementing rapid susceptibility tests.