Prevalence and antimicrobial resistance of beta-lactamase-producing Gram-negative isolates from outpatient clinical and environmental samples in the Zenica-Doboj Canton , Bosnia and Herzegovina

Introduction: Infections due to extended-spectrum beta-lactamase (ESBL)-producing isolates in patients are hard to treat and cause high morbidity and mortality. ESBL-producing bacteria have been increasingly detected in environmental samples in different countries since 2002, and have gained considerable attention worldwide. Methods: Antibiotic susceptibility of all isolates was determined using the disk diffusion method. The production of ESBLs was determined by the double-disk synergy test. Results: Among the outpatient clinical samples, out of 2857 Gram-negative bacteria, 184 (6.5%) ESBLproducing bacteria were isolated. In this group, 143 (77.7%) were from urine samples, 26 (14.1%) from surgical wounds, 6 (3.3%) from umbilical swabs, and 9 (4.9%) from other patients sites (upper respiratory tract, cannula, eyes, genital swabs). Escherichia coli was isolated in 62 (33.7%), and Klebsiella spp. in 50 (27.8%) cases. Among the environmental samples, out of 381 Gram-negative bacteria, 52 (13.6%) were ESBL-producing isolates. In this group, 37 (71.2%) were sampled from water, 7 (13.5%) from food, and 8 (15.4%) from environmental surfaces. The most prevalent ESBL-producing bacteria isolated from the environmental samples were E. coli (isolated from 26 samples), Klebsiella spp. (10), non-fermenters (9), and other bacteria isolated from 7 samples. The clinical outpatient ESBL-producing isolates showed resistance to all cephalosporins, ranging from 25% (cefepime) to 100% (cefuroxime). The environmental ESBL-producing isolates showed resistance to cefuroxime, aztreonam, cefpodoxime, amoxicillin/clavulanate, and cefoxitin in the range of 65-100%. Conclusions: Prevalence of antibiotic resistance of ESBL-producing strains is high and requires routine detection of ESBL-producing isolates in the laboratories, designing of appropriate antibiotic prescribing policies and control of the risk factors.


INTRODUCTION
Over the past decades, antimicrobial resistance has been recognized as an important global health problem in many fields, such as human and veterinary medicine, livestock holdings, agriculture, and environment (1,2).In countries with low social and economic status, especially countries in South and Southeast Asia, antimicrobial resistance is more prevalent (3).The production of extended-spectrum beta-lactamases (ESBLs) or AmpC beta-lactamases is one of the most important mechanisms of resistance to extended-spectrum cephalosporins in the Gram-negative isolates (4).ESBLs are plasmid encoded enzymes and their genes are usually transported by plasmids.The enzymes are capable of inactivating a large number of beta-lactam antibiotics, including an extended spectrum and a very broad spectrum cephalosporins and monobactams.The overexpression of chromosomal or plasmid-mediated AmpC enzymes in patients can cause resistance to broad spectrum cephalosporins (5).
In the past decades, the ESBLs were mostly associated with hospitals and institutional care in humans, but they are now increasingly found in the community, in food producing animals, and in water (6)(7)(8).They have been detected in food for animals at farms, meat, water, and the environment in many countries such as Italy, China, Malaysia, Nigeria, Austria, and the Netherlands (5,(9)(10)(11)(12)(13)(14).
Investigations of prevalence and antibiotic resistance in Bosnia and Herzegovina (B&H) are scarce.ESBLs and AmpC beta-lactamases have been reported in isolates causing urinary tract infections and nosocomial infections in the Zenica-Doboj Canton and Tuzla Canton (6,15).
The aim of this study was to investigate the prevalence and antibiotic resistance of ESBL-producing isolates in the Zenica-Doboj Canton, B&H.

Antimicrobial susceptibility testing
The susceptibility testing to 13 antimicrobials was performed by a two-fold microdilution technique according to the Clinical and Laboratory Standards Institute (CLSI) procedure (16).Escherichia coli ATCC 25922 (ESBL negative) and Klebsiella pneumoniae 700603 (ESBL positive) were used as quality control strains.

Phenotype detection of ESBLs
ESBL production was determined by the double disk-synergy test (DDST).Overnight broth culture of the test strain was diluted in saline, adjusted to McFarland standard suspension 0.5 and inoculated onto Mueller-Hinton agar (MHA); disk containing amoxicillin/clavulanate (20/10 µg) was placed in the middle of the plate and surrounded (20 mm distance center to center) by disks containing cefotaxime (5 µg), ceftriaxone (30 µg), ceftazidime (10 µg), and cefepime (30 µg) (Becton-Dickinson, USA).Plates were incubated overnight at 37 °C.Any distortion or increase of the inhibition zones around cephalosporin disks toward amoxicillin/ clavulanate disk was indicative of ESBL production (16).Production of ESBLs was confirmed by CLSI combined disk test.Disks containing 30 µg of cefotaxime and ceftazidime, and disks containing a combination of the two drugs plus 10 µL (10 µg) of clavulanic acid (Becton-Dickinson, USA) were placed independently, 20 mm apart, on a lawn culture of 0.5 McFarland opacity of the test isolate on the Mueller-Hinton agar plate and incubated for 18-24 hours at 35°C.Isolates were considered ESBL positive if the inhibition zone measured around one of the combination disks after the overnight incubation was at least 5 mm larger than that of the corresponding cephalosporin disk (16).Isolates resistant to the extended-spectrum cephalosporins and β-lactam/β-lactamase inhibitor combination (amoxicillin/clavulanic acid) were screened for the production of AmpC β-lactamases by the combined disk test using 3-amino phenylboronic acid (PBA) (Sigma-Aldrich, Steinheim, Germany).The stock solution was prepared as previously recommended ( 17) by dissolving PBA (benzeneboronic acid; Sigma-Aldrich, Steinheim, Germany) in dimethyl sulfoxide at a concentration of 20 mg/mL.Twenty µL (containing 400 µg of boronic acid) of the solution was dispensed onto antibiotic disks.The disks were then dried and used within 60 min.The tests were performed by inoculating Mueller-Hinton agar by the standard diffusion method and placing disks containing four different β-lactams (CAZ, 10 µg; CRO, 30 µg; CTX, 5 µg; FEP, 30 µg) with or without boronic acid.The agar plates were incubated at 37°C overnight.The diameter of the growth-inhibitory zone around a β-lactam disk with boronic acid was compared to the zone around the corresponding β-lactam disk without boronic acid.The test was considered positive for the detection of AmpC production when the diameter of the growth-inhibitory zone around the β-lactam disk with boronic acid was ≥5 mm larger than the diameter around the disk without boronic acid (18).Production of carbapenemases of the class A or class B was confirmed by the combined disk-test using meropenem disks with PBA and EDTA (ethylenediaminetetraacetic acid) (Sigma-Aldrich, Steinheim, Germany), respectively (19).Three meropenem (MEM) disks were placed on Mueller-Hinton agar plate inoculated with the test strain.Ten µL of EDTA (300 mg) and PBA (300 mg) was added on the first and third disks, respectively.The difference of ≥5 mm in the zone size between the disks with and without EDTA was suggesting the production of carbapenemase class B, and the difference of ≥5 mm in the zone size between the disks with and without PBA was suggesting the production of carbapenemase class A (19).

Antibiotic susceptibility patterns of the isolates from environmental samples
Table 2 demonstrates the antimicrobial resistance results of 52 phenotypically positive beta-lactamase-producing isolates, out of which 26 were E. coli isolates, 10 Klebsiella spp., 9 non-fermenters, and 7 were from other isolates (4 Citrobacter spp.and 3 Enterobacter spp.).
Cephalosporins resistance rates were high in Klebsiella spp., following were the rates in E. coli.The bacteria showed 20 -100% overall resistance rates, and all Klebsiella spp.isolates were susceptible to imipenem and meropenem (Table 2).
Non-fermenters and other isolates showed higher resistance rates to the 3 rd and 4 th generations of cephalosporins compared to E. coli and Klebsiella spp.(Table 2).

DISCUSSION
This study aimed to determine variability in the prevalence of ESBL-producing isolates causing infections in humans, and ESBL-producing isolates from environmental samples such as food, water, and environmental surfaces.
Detection of the ESBL production was performed with the combined disk method and showed that 6.5% of the clinical isolates were considered as potential producers of ESBLs, which was lower than in the reports from Serbia, the Republic of Macedonia, Turkey, and Egypt (20,21).However, our results are similar to those reported for Austria, Croatia, Slovenia, Japan, and Tunis (20,22,23).
The phenotypic cefoxitin resistance test showed a higher prevalence (70%) of the production of AmpC beta-lactamases which is contradictory to the report from Morocco (24), but similar to the report from the  Netherlands (25).The number of potential producers of ESBLs and AmpC beta-lactamases in the isolates from the environmental samples was higher (13.6%) than in the isolates from the clinical outpatient samples, which is in agreement with the report from India (26).Patternel et al. reported a higher prevalence of ESBL-producing E. coli isolates from minced meat in 24% cases, in Graz, Austria, (13).Similarly, investigators from the Netherlands reported 55% of ESBLproducing E. coli isolates collected from retail chicken meat and surface water (14) and 100% from wastewater samples (27), which is contradictory to our report.
The ESBL-producing outpatient clinical isolates showed high rates of resistance to all cephalosporins, except to cefepime and high rates of resistance to aminoglycosides and fluoroquinolones, similar to the reports from India and Pakistan (28,29).In 2014, the World Health Organization reported that E. coli/K.pneumoniae was resistant to the 3 rd generation of cephalosporins in 9.1/13.3%cases (Austria), 21.3/82.1% (Serbia), 8.8/30.2%(Slovenia), and 47.4/91% (the Republic of Macedonia) among others (20).In some instances, these results are similar to ours, however there are some differences as well.For the treatment of these infections in the community, first-line antibiotics were gentamicin or ciprofloxacin (cheap antibiotics without medical prescription), and this is similar to the report from Pakistan (29).
All isolates from our environmental samples showed high resistance rates to cefoxitin, cefuroxime, aztreonam, and amoxicillin/clavulanic acid, but low resistance to the 3 rd and 4 th generations of cephalosporins, which is also reported in Egypt (21).
Resistant bacteria can cause foodborne diseases, at first a silent carrier state but later may cause infections that are not recognized as being of foodborne origin.ESBL-producing isolates could be transmitted by the food production chain and environmental samples (30).
Boonyasiri et al. reported that basic sanitation and hygiene, including consumption of cooked food and clean water, as well as hand washing, are very important measures for containment and prevention of antibiotic resistance (30).
The main limitation of this study is the small number of ESBL-producing isolates that were collected/available for the analysis, because of the short time span (six months) and due to analysis of only one region-Canton, out of ten cantons in Bosnia and Herzegovina.It is not representative for whole B&H, because we have a different number of populations in each canton, different lifestyles with different social conditions, different climate, etc.However, reduced susceptibility of these isolates is a worldwide concern.Because of the high prevalence of the resistant ESBL-producing strains in the epidemiologically unrelated patients and environmental samples in this study, further local surveillance is needed.Molecular characterization, or at least phenotypic testing of the ESBL production of Gram-negative bacteria, is important for appropriate therapy and the detection of the sources and modes of the spread, which is further the main step in designing targeted infection control strategies.
Between December 2009 and May 2010, and between December 2013 and May 2014, a random sample of 184 clinically relevant and non-duplicate isolates was routinely collected in the Microbiology Laboratory at the Cantonal Hospital Zenica.The Cantonal Hospital Zenica is 849 bed tertiary level hospital admitting about 25 000 patients/year, with 240 000 hospital days, and covers a population of 331 229 in the Zenica-Doboj Canton.