Fluoroquinolone resistance in Escherichia coli and Klebsiella pneumoniae over 18 years: effect of different systems for eliminating duplicates

Abstract

Sir, The increase in resistance of Gram-negative bacilli is a serious public health problem and our aim was to determine the effect of different duplicate elimination criteria on the evolution of fluoroquinolone resistance in Escherichia coli and Klebsiella pneumoniae in our setting over a period of 18 years. The criteria used are described in Figure 1. Antibiotic susceptibility was studied using microdilution with the Wider semiautomatic system (Soria Melguizo, Spain) and Etest (AB Biodisk, Solna, Sweden), following CLSI (formerly NCCLS) recommendations. In E. coli, the mean percentage of fluoroquinolone-resistant strains according to the EARS-Net criterion was greater than that detected using the CLSI criterion, both when the total number of strains and when the number of extended-spectrum b-lactamase (ESBL)-producing strains was considered (28.4% versus 22.4% and 65.2% versus 60.5%, respectively). With regard to K. pneumoniae, we obtained 7.2% versus 6.3% and 52.2% versus 40.0%. Analysis of the criterion of time showed that when the period considered for elimination of duplicates increased, the ciprofloxacin resistance rate also increased compared with the CLSI criterion. When the EARS-Net criterion was applied, the prevalence of these microorganisms increased in both hospitalized and non-hospitalized patients, although this phenomenon was more important in the former than in the latter. All data are shown in Table 1. The increase in fluoroquinolone resistance is a serious public health problem, especially in ESBL-producing strains, for which the therapeutic options are far more limited, and the local epidemiology of resistance has been considered a key tool in the control of this process. It is essential to apply a duplicate elimination criterion, since analysis of the data based on the total criterion does not reflect the real situation of resistance. What has not yet been definitively established is which criterion is the best to use in each case. The first isolate from each patient is an objective criterion, but has serious limitations since the same patient may frequently be infected by different strains of the same species during the course of their disease and hospital stay, or a strain may become resistant during treatment of the patient due to the selection of resistant mutants during antibiotic therapy. The EARS-Net criterion reflects more faithfully the real situation of resistance, but suggests that it is necessary to apply strict quality control in phenotypic microbiological studies to detect resistance. Our study suggests that the new EARS-Net method of studying duplicates should be applied in order to detect changes in resistance patterns that may occur during the course of disease due to mutations of the microorganism causing the infection or superinfections by more resistant microorganisms.