Epidemiological cut-off values for Yersina ruckeri disc diffusion data generated by a standardised method.

In order to establish the meaning of data generated in antimicrobial agent susceptibility tests, it is necessary to develop internationally harmonised interpretive criteria. Currently, such criteria have not been developed for data generated in studies of the susceptibility of the fish pathogen Yersinia ruckeri. This work generated the data that would be required to set epidemiological cut-off values for the susceptibility data of this species that had been generated using a standardised disc diffusion method that specified the use of Mueller Hinton agar and incubation at 22°C for 24-28 h. Using this method, sets of inhibition zones data for 4 antimicrobial agents were generated by 3 independent laboratories. The data from these laboratories were aggregated and analysed using the statistically based normalised resistance interpretation. For ampicillin, florfenicol, oxytetracycline and trimethoprim-sulfamethoxazole the cut-off values calculated by this analysis were ≥16, ≥23, ≥24 and ≥30 mm, respectively. Evidence is presented demonstrating that the data for these 4 agents was of sufficient quantity and quality that they could be used by the relevant authorities to set internationally harmonised, consensus epidemiological cut-off values for Y. ruckeri.

Epidemiological cut-off values for non-O1/ non-O139 Vibrio cholerae disc diffusion data generated by standardised methods.

This work generates the data needed to set epidemiological cut-off values for disc-diffusion zone measurements of Vibrio cholerae. The susceptibility of 147 European isolates of non-O1/non-O139 V. cholerae to 19 antibiotics was established using a standardised disc diffusion method which specified incubation of Mueller Hinton agar plates at 35°C. Epidemiological cut-off values were calculated by analysis of the zone size data with the statistically based normalised resistance interpretation method. Cut-off values for 17 agents were calculated by analysis of the aggregated data from all 4 laboratories participating in this study. The cut-off values calculated were ≥18 mm for amoxicillin/clavulanate, ≥18 mm for amikacin, ≥19 mm for ampicillin, ≥27 mm for cefepime, ≥31 mm for cefotaxime, ≥24 mm for ceftazidime, ≥24 mm for chloramphenicol, ≥31 mm for ciprofloxacin, ≥16 mm for erythromycin, ≥ 27 mm for florfenicol, ≥16 mm for gentamicin, ≥23 mm for imipenem, ≥25 mm for meropenem, ≥29 mm for nalidixic acid, ≥28 mm for norfloxacin, ≥13 mm for streptomycin and ≥23 mm for tetracycline. For the other 2 agents the data from 1 laboratory was excluded from the censored aggregation because the data from that laboratory was considered excessively imprecise. The cut-off values for these 2 agents calculated for the aggregation of the data from 3 laboratories were ≥23 mm for trimethoprim and ≥24 mm for trimethoprim/sulfamethoxazole. These zone size data will be submitted to the Clinical Laboratory Standards Institute (CLSI) and European Committee for Antimicrobial Susceptibility Testing (EUCAST) for their consideration in setting international consensus epidemiological cut-off values for non O1/non-O139 V. cholerae.

Epidemiological cut-off values for Vibrio anguillarum MIC and disc diffusion data generated by standardised methods.

This work aims to generate the data needed to set epidemiological cut-off values for minimum inhibitory concentration (MIC) and disc-diffusion zone measurements of Vibrio anguillarum. A total of 261 unique isolates were tested, applying standard methods specifying incubation at 28°C for 24-28 h. Aggregated MIC distributions for a total of 247 isolates were determined in 9 laboratories for 11 agents. Data aggregations of the disc zone for the 10 agents analysed contained between 157 and 218 observations made by 4 to 7 laboratories. Acceptable ranges for quality control (QC) reference strains were available for 7 agents and the related multi-laboratory aggregated data were censored, excluding the data of a laboratory that failed to meet QC requirements. Statistical methods were applied to calculate epidemiological cut-off values. Cut-off values for MIC data were calculated for florfenicol (≤1 µg ml-1), gentamicin (≤4 µg ml-1), oxytetracycline (≤0.25 µg ml-1) and trimethoprim/sulfamethoxazole (≤0.125/2.38 µg ml-1). The cut-off values for disc zone data were calculated for enrofloxacin (≥29 mm), florfenicol (≥27 mm), gentamicin (≥19 mm), oxolinic acid (≥24 mm), oxytetracycline (≥24 mm) and trimethoprim/sulfamethoxazole (≥26 mm). MIC and disc-diffusion zone data for the other agents where not supported by QC, thus yielding only provisional cut-off values (meropenem, ceftazidime). Regardless of whether QC is available, some of the aggregated MIC distributions (enrofloxacin, oxolinic acid), disc zone (sulfamethoxazole), and MIC and disc-diffusion distributions (ampicillin, chloramphenicol) did not meet the statistical requirements. The data produced will be submitted to the Clinical Laboratory Standards Institute for their consideration in setting international consensus epidemiological cut-off values.

Abundance and environmental host range of the SXT/R391 ICEs in aquatic environmental communities.

Mobile genetic elements (MGEs) such as plasmids or integrative conjugative elements (ICEs) are widely involved in the horizontal transfer of antibiotic resistant genes (ARGs), but their environmental host-range and reservoirs remain poorly known, as mainly assessed through the analysis of culturable and clinical bacterial isolates. In this study, we used a gradual approach for determining the environmental abundance and host-range of ICEs belonging to the SXT/R391 family, otherwise well known to bring ARGs in Vibrio spp. epidemic clones and other pathogens. First, by screening a set of aquatic bacteria libraries covering 1794 strains, we found that almost 1% of the isolates hosted an SXT/R391 element, all belonging to a narrow group of non-O1/non-O139 Vibrio cholerae. However, when SXT/R391 ICEs were then quantified in various aquatic communities, they appeared to be ubiquitous and relatively abundant, from 10-6 to 10-3 ICE copies per 16 S rDNA. Finally, the molecular exploration of the SXT/R391 host-range in two river ecosystems impacted by anthropogenic activities, using the single-cell genomic approach epicPCR, revealed several new SXT/R391 hosts mostly in the Proteobacteria phylum. Some, such as the pathogen Arcobacter cryaerophilus (Campylobacteraceae), have only been encountered in discharged treated wastewaters and downstream river waters, thus revealing a likely anthropogenic origin. Others, such as the non-pathogenic bacterium Neptunomonas acidivorans (Oceanospirillaceae), were solely identified in rivers waters upstream and downstream the treated wastewaters discharge points and may intrinsically belong to the SXT/R391 environmental reservoir. This work points out that not only the ICEs of the SXT/R391 family are more abundant in the environment than anticipated, but also that a variety of unsuspected hosts may well represent a missing link in the environmental dissemination of MGEs from and to bacteria of anthropogenic origin.

Agreement between the categorization of isolates of Aeromonas salmonicida and Yersinia ruckeri by disc diffusion and MIC tests performed at 22℃.

Standard disc diffusion and MIC test procedure were used to investigate the susceptibility of two hundred and fifty-one isolates collected from infected fish in France to florfenicol, oxolinic acid and tetracycline. The tests were performed at 22 ± 2℃ and for the 177 Yersinia ruckeri they were read after 24-28 hr incubation and for the 74 Aeromonas salmonicida isolates they were read after 44-48 hr. Applying epidemiological cut-off values to the susceptibility data generated in these tests, the isolates were categorized as wild-type or non-wild-type. The agent-specific categories into each isolate were placed on the basis of the data generated by the two methods were in agreement in 98% of the determinations made. It is argued that, with respect to categorising isolates, disc diffusion and MIC methods can be considered as equally valid at this temperature and after both periods of incubation.

Characterisation of plasmids harbouring extended-spectrum cephalosporin resistance genes in Escherichia coli from French rivers.

Antimicrobial resistance is a "One Health" issue that requires improved knowledge of the presence and abundance of resistant bacteria in the environment. Extended-spectrum cephalosporins (ESCs) are critically important antibiotics (CIAs), and resistance to these CIAs is often encoded by beta-lactamase genes borne on conjugative plasmids. We thus decided to characterise 21 plasmids of ESC-resistant Escherichia coli randomly selected from isolates previously obtained from river water collected in a rural area in western France. The plasmids encoding ESC resistance were sequenced to investigate the diversity of the genes encoding ESC resistance and their genetic context. Sequences revealed that eleven IncI1 pMLST3 plasmids carried the blaCTX-M-1 and sul2 genes, and some of them also had the tet(A), aadA5 or dfrA17 genes. The blaCTX-M-1 gene was also detected on an IncN plasmid. Five plasmids obtained from four rivers contained blaCTX-M-14, either on IncI1 or on IncFII plasmids. Two strains from two rivers contained blaCTX-M-15 on IncN pMLST7 plasmids, with qnrS1 and dfrA14 genes. One plasmid contained the blaCTX-M-55, a blaTEM-1B-like, and fosA genes. One plasmid contained the blaCMY-2 gene. The diversity of the genes and plasmids of the resistant bacteria isolated from French rivers is probably related to the various animal and human origins of the isolated bacteria.

Phenotypic and Genotypic Antimicrobial Resistance Traits of Vibrio cholerae Non-O1/Non-O139 Isolated From a Large Austrian Lake Frequently Associated With Cases of Human Infection.

Vibrio cholerae belonging to serogroups other than O1 and O139 are opportunistic pathogens which cause infections with a variety of clinical symptoms. Due to the increasing number of V. cholerae non-O1/non-O139 infections in association with recreational waters in the past two decades, they have received increasing attention in recent literature and by public health authorities. Since the treatment of choice is the administration of antibiotics, we investigated the distribution of antimicrobial resistance properties in a V. cholerae non-O1/non-O139 population in a large Austrian lake intensively used for recreation and in epidemiologically linked clinical isolates. In total, 82 environmental isolates - selected on the basis of comprehensive phylogenetic information - and nine clinical isolates were analyzed for their phenotypic antimicrobial susceptibility. The genomes of 46 environmental and eight clinical strains were screened for known genetic antimicrobial resistance traits in CARD and ResFinder databases. In general, antimicrobial susceptibility of the investigated V. cholerae population was high. The environmental strains were susceptible against most of the 16 tested antibiotics, except sulfonamides (97.5% resistant strains), streptomycin (39% resistant) and ampicillin (20.7% resistant). Clinical isolates partly showed additional resistance to amoxicillin-clavulanic acid. Genome analysis showed that crp, a regulator of multidrug efflux genes, and the bicyclomycin/multidrug efflux system of V. cholerae were present in all isolates. Nine isolates additionally carried variants of bla CARB-7 and bla CARB-9, determinants of beta-lactam resistance and six isolates carried catB9, a determinant of phenicol resistance. Three isolates had both bla CARB-7 and catB9. In 27 isolates, five out of six subfamilies of the MATE-family were present. For all isolates no genes conferring resistance to aminoglycosides, macrolides and sulfonamides were detected. The apparent lack of either known antimicrobial resistance traits or mobile genetic elements indicates that in cholera non-epidemic regions of the world, V. cholerae non-O1/non-O139 play a minor role as a reservoir of resistance in the environment. The discrepancies between the phenotypic and genome-based antimicrobial resistance assessment show that for V. cholerae non-O1/non-O139, resistance databases are currently inappropriate for an assessment of antimicrobial resistance. Continuous collection of both data over time may solve such discrepancies between genotype and phenotype in the future.

Single-strain regression analysis evaluating disc potencies of flumequine and enrofloxacin for testing Aeromonas sobria and Vibrio anguillarum.

Two quite different disc contents are used for antimicrobial susceptibility testing of two fluoroquinolone drugs, flumequine and enrofloxacin, in the disc diffusion test, 30 and 5 µg, respectively. Using the SRA method, single-strain regression analysis, we studied the impact of disc content when testing two relevant bacterial species, Aeromonas sobria and Vibrio anguillarum. There were no major differences between the antimicrobial regression lines for the two species. Wild-type strains produced acceptable zones of inhibition over a wide range of disc contents. The flumequine 30 µg disc should be lowered in its drug content. No rational reasons for choosing so different disc contents for the two antimicrobials were apparent. At present, the choice of disc content for new antimicrobials are outside the realm of clinical microbiologists. It is recommended that reference authorities, such as EUCAST, CLSI and USCAST, are consulted for the choice of disc contents in the future.

Longitudinal study of Escherichia coli plasmid resistance to extended-spectrum cephalosporins in free-range broilers.

Resistance to extended-spectrum cephalosporins (ESCs) is mostly borne by conjugative plasmids. The aim of the present study was to evaluate the characteristics and diversity of ESC resistance plasmids in Escherichia coli from different free-range broiler flocks in France, and their persistence in flocks during rearing. Two hatcheries were selected. Faecal samples from 11 flocks were collected from before their arrival on the broiler production farm up to their slaughter at the end of the rearing period. A selection of 25 E. coli isolates obtained at different times from different flocks but all harbouring an ESC resistance gene was characterised. The plasmids coding for ESC resistance were sequenced using Mi-seq Illumina technology or the ion proton system (Ion Torrent). Ten IncI1 ST12 plasmids carried the blaCMY-2 gene, and most of them had no other resistance genes. All blaCMY-2 plasmids were obtained from day-old to 7-day-old chicks from four flocks hatched at the same hatchery and sent to three different farms. Sequence comparisons showed identity percentages higher than 99%. Fifteen IncI1 ST3 plasmids were obtained from day-old to 77-day-old broilers from seven flocks on six farms. These plasmids harboured the blaCTX-M-1 gene, and most also had the tet(A) and sul2 genes, with sequence identity higher than 99%. For both types of plasmid, very high identity percentages were also obtained with published sequences of plasmids isolated from broilers in other countries or from other animal species. Thus, unlike the IncI1 ST12 blaCMY-2 plasmids, the epidemic nature of the IncI1 ST3 blaCTX-M-1 plasmids in the French poultry production makes it difficult to determine the origin of a contamination which may persist for weeks in a flock.

Antimicrobial Susceptibility among Urban Wastewater and Wild Shellfish Isolates of Non-O1/Non-O139 Vibrio cholerae from La Rance Estuary (Brittany, France).

The early 2000s marked the end of the Golden age of the antibiotics and the beginning of the awareness on the potential threat to human health due to the dissemination of antimicrobial resistance. As a base-line study, we investigated the antimicrobial susceptibility of 99 strains of non-O1/non-O139 Vibrio cholerae isolated from wastewater and shellfish in 2000/2001 within La Rance estuary (Brittany, France). All isolates were susceptible to amoxicillin-clavulanic acid, cefotaxime, imipenem, chloramphenicol, nalidixic acid, ciprofloxacin, norfloxacin, amikacin, gentamicin, tetracycline, doxycycline, trimethoprim-sulfamethoxazole, and erythromycin. The only resistances were to streptomycin, sulfonamides and ampicillin: 54.6% of the isolates had acquired resistance to at least one antimicrobial agent among them and only six isolates from cockles were multidrug resistant. On the basis of the distribution of a limited selection of resistance associated genes, our study shows that V. cholerae can constitute an environmental reservoir for these genes. However, none of our isolates harbored integron. This result casts doubt on the capacity of non-O1/non-O139 V. cholerae to acquire resistance-associated genes in such context, and on its potential role of indicator of the dissemination of antimicrobial resistance in the aquatic environment.

Aeromonas Diversity and Antimicrobial Susceptibility in Freshwater-An Attempt to Set Generic Epidemiological Cut-Off Values.

The importance of the role of environment in the dissemination of antimicrobial resistant bacteria is now well recognized. Thus, bacterial indicators to monitor the phenomena are required. The Aeromonas genus is autochthonous in the aquatic environment and easy to detect in any water type, such as freshwater, or wastewater. These microorganisms are also causing infections in humans and animals (including fish). Furthermore, as Aeromonas spp. is able to acquire antimicrobial resistance mechanisms, it is candidate for indicator bacteria to follow antimicrobial resistance dissemination in aquatic environments. Unfortunately, to date, interpretation criteria for Aeromonas spp. for antimicrobial susceptibility tests are scarce in the literature. No epidemiological cut-off values for Aeromonas are currently available at EUCAST to interpret Minimum Inhibitory Concentrations (MIC). The only interpretation criteria available are clinical breakpoints from CLSI that are adapted from Enterobacteriaceae. Based on the results of MIC distributions obtained for a collection of environmental isolates of Aeromonas, this study aimed at proposing tentative epidemiological cut-off values (COWT) for Aeromonas spp. assessing whether the genus is an acceptable level of definition. Thus, 233 isolates collected from 16 rivers were identified at species level using Maldi-Tof (Bruker). Eleven different species were identified, the most abundant were A. bestiarum (n = 54), A. salmonicida (n = 45), A. sobria (n = 41), and A. eucrenophila (n = 37). 96-well micro-plates containing different concentrations of 15 antimicrobials, namely cefotaxime, ceftazidime, chloramphenicol, colistin, enrofloxacin, erythromycin, florfenicol, flumequine, gentamicin, nalidixic acid, oxolinic acid, streptomycin, temocillin, tetracycline, and trimethoprim-sulfamethoxazole, were prepared. The broth micro-dilution method was used to determine the antimicrobial susceptibility of each isolate. The estimation of COWT values was satisfactory obtained at genus level for all antimicrobials except cefotaxime and erythromycin. This first step is an invitation for other research teams to increase the amount of antimicrobial resistance data collected. Then, robustness of our proposed provisional generic epidemiological cut-off values could be assessed by testing antimicrobial susceptibility of various Aeromonas collections.

Antimicrobial Susceptibility of Autochthonous Aquatic Vibrio cholerae in Haiti.

We investigated the antimicrobial susceptibility of 50 environmental isolates of Vibrio cholerae non-O1/non-O139 collected in surface waters in Haiti in July 2012, during an active cholera outbreak. A panel of 16 antibiotics was tested on the isolates using the disk diffusion method and PCR detection of seven resistance-associated genes (strA/B, sul1/2, ermA/B, and mefA). All isolates were susceptible to amoxicillin-clavulanic acid, cefotaxime, imipenem, ciprofloxacin, norfloxacin, amikacin, and gentamicin. Nearly a quarter (22.0%) of the isolates were susceptible to all 16 antimicrobials tested and only 8.0% of the isolates (n = 4) were multidrug-resistant. The highest proportions of resistant isolates were observed for sulfonamide (70.0%), amoxicillin (12.0%), and trimethoprim-sulfamethoxazole (10.0%). One strain was resistant to erythromycin and one to doxycycline, two antibiotics used to treat cholera in Haiti. Among the 50 isolates, 78% possessed at least two resistance-associated genes, and the genes sul1, ermA, and strB were detected in all four multidrug-resistant isolates. Our results clearly indicate that the autochthonous population of V. cholerae non-O1/non-O139 found in surface waters in Haiti shows antimicrobial patterns different from that of the outbreak strain. The presence in the Haitian aquatic environment of V. cholerae non-O1/non-O139 with reduced susceptibility or resistance to antibiotics used in human medicine may constitute a mild public health threat.

Impact of the administration of a third-generation cephalosporin (3GC) to one-day-old chicks on the persistence of 3GC-resistant Escherichia coli in intestinal flora: An in vivo experiment.

The aim of the experiment was to evaluate under controlled conditions the impact on the excretion of 3GC-resistant Escherichia coli of the injection of one-day-old chicks with ceftiofur, a third-generation cephalosporin (3GC). Three isolators containing specific-pathogen-free chicks were used. In the first one, 20 birds were injected with ceftiofur then ten of them were orally inoculated with a weak inoculum of a 3GC-resistant E. coli field isolate containing an IncI1/ST3 plasmid encoding a blaCTX-M-1 beta-lactamase. The other chicks were kept as contact birds. None of the 20 birds in the second isolator were injected with ceftiofur, but ten of them were similarly inoculated with the 3GC-resistant strain and the others kept as contact birds. A third isolator contained ten non-injected, non-inoculated chicks. Fecal samples were collected regularly over one month and the E. coli isolated on non-supplemented media were characterized by antimicrobial agar dilution, detection of selected resistance genes and determination of phylogenetic group by PCR. The titers of 3GC-resistant E. coli in individual fecal samples were evaluated by culturing on 3GC-supplemented media. Results showed that the inoculated strain rapidly and abundantly colonized the inoculated and contact birds. The ceftiofur injection resulted in significantly higher percentages of 3GC-resistant E. coli isolates among the analyzed E. coli. No transfer of the 3GC-encoding plasmid to other isolates could be evidenced. In conclusion, these results highlight the dramatic capacity of 3GC-resistant E. coli to colonize and persist in chicks, and the selecting pressure imposed by the off-label use of ceftiofur.

Impact of third-generation-cephalosporin administration in hatcheries on fecal Escherichia coli antimicrobial resistance in broilers and layers.

We investigated the impact of the hatchery practice of administering third-generation cephalosporin (3GC) on the selection and persistence of 3GC-resistant Escherichia coli in poultry. We studied 15 3GC-treated (TB) and 15 non-3GC-treated (NTB) broiler flocks and 12 3GC-treated (TL) and 10 non-3GC-treated (NTL) future layer flocks. Fecal samples from each flock were sampled before arrival on the farm (day 0), on day 2, on day 7, and then twice more. E. coli isolates were isolated on MacConkey agar without antibiotics and screened for 3GC resistance, and any 3GC-resistant E. coli isolates were further analyzed. 3GC-resistant E. coli isolates were found in all 3GC-treated flocks on at least one sampling date. The percentages of 3GC-resistant E. coli isolates were significantly higher in TB (41.5%) than in NTB (19.5%) flocks and in TL (49.5%) than in NTL (24.5%) flocks. In the day 2 samples, more than 80% of the E. coli strains isolated were 3GC resistant. 3GC-resistant E. coli strains were still detected at the end of the follow-up period in 6 out of 27 3GC-treated and 5 out of 25 non-3GC-treated flocks. Many 3GC-resistant E. coli strains were resistant to tetracycline, and there were significant differences in the percentages of resistance to sulfamethoxazole-trimethoprim, streptomycin, or gentamicin between treated and nontreated flocks. blaCTX-M and blaCMY-2 were the most frequently detected genes. These results clearly demonstrated that 3GC-resistant strains are introduced early in flocks and that the use of 3GC in hatcheries promotes the selection of 3GC-resistant E. coli. Measures must be implemented to avoid the spread and selection of 3GC-resistant strains.