Evaluation of the susceptibility interpretation on Escherichia coli, Klebsiella pneumonia, Proteus mirabilis in China by agar dilution method according to the changes of cephalosporin breakpoints in CLSI 2010 / 中华检验医学杂志

ABSTRACT

Objective To evaluate the influences of susceptibility interpretation of Escherichia coli,Klebsiella pneumonia and Proteus mirabilis in China mainland according to the old and new ceftazidime,cefotaxime and ceftriaxone breakpoints in CLSI M100-S20 and CLSI M100-S19. Methods First, We analyzed the antibacterial susceptibility results of the three bacteria by agar dilution method in the SEANIR surveillance item, which were collected from 15 national hospitals between the year of 2005 and 2007 and excluded the AmpC enzyme positive isolates according to the PGR-DNA sequencing method and/or the antibacterial susceptibility phenotype. ESBL phenotype was confirmed by the CLSI phenotypic confirmatory test. Antibacterial susceptibility of the total 2733 Escherichia coli, Klebsiella pneumonia, Proteus mirabilis isolates was retrospectively analyzed by WHONET 5. 4 software according to the breakpoints of the CLSI M100-S19 (S19) and CLSI M100-S20 (S20). Second, 207 isolates of Peking Union Medical College Hospital with the results of both agar dilution method and disk diffusion method were performed by recurrent analysis. Then we observed the inter-method agreement through the scatter diagram according to the breakpoints of S19 and S20. Results First, as to the ESBL positive Escherichia coli, Klebsiella pneumonia and Proteus mirabili.s, the resistant rate of cefotaxime increased from 65.2% , 55.5%, 14. 6% under S19 (64 μg/ml) to 99. 7%, 96. 2% , 93. 8% under S20 (4 μg/ml). The susceptibility rates decreased from 6. 0%, 11.5%, 33.3% under S19 (8 μg/ml) to 0%, 0. 2%, 0% under S20 ( 1 μg/ml). Ceftriaxone had the same trend as cefotaxime. Though ceftazidime was more active than cefotaxime and ceftriaxone, as to the ESBL positive Escherichia coli and Klebsiella pneumonia, the resistant rates slightly increased from 30. 3%,43. 2% under S19 (32 μg/ml) to42.0%, 56. 0% under S20 (16 μg/ml). The susceptibility rates slightly decreased from 58. 1%, 44. 1% under S19 (8 μg/ml) to 44. 7%, 28.0% under S20 (4 μg/ml). Second,as to the ESBL negative Escherichia coli, Klebsiella pneumonia and Proteus mirabilis, all the susceptibility rates of ceftazidime, cefotaxime and ceftriaxone were between 99. 2%-100. 0%, the resistant rate were between 0%-0. 4%. Third, the S20 MIC breakpoints had a good correspondence with the ESBL phenotype.Fourth, according to the recurrent analysis of MIC testing and disk dilution method, r value was 0. 67,0. 79, 0. 77 for ceftazidime, cefotaxime and ceftriaxone, respectively, and all P value were under 0. 01. The intermethod rates of S19 and S20 were both acceptable. Conclusions If the cefotaxime and ceftriaxone S20 new breakpoints were used, the concordance of antibacterial susceptibility results and ESBL phenotype would increase greatly. The clinician could select proper antibiotics according to the antibacterial susceptibility results and clinical symptoms. It is no longer necessary to edit results for cephalosporins, aztreonam, or penicillins from susceptible to resistant. However, until laboratories implement the new interpretive criteria,ESBL testing should be performed as described in Supplemental Table 2A-S1. The relationship between the new breakpoints of ceftazidime and clinical outcomes need to be further evaluated.