Background
Enterobacteriaceae carrying extended-spectrum β-lactamases (ESBLs) is a global concern that demands global attention due to the limited available treatment options [
1‐
4]. Over the past two decades, there has been an exponential increase in β-lactamase resistance worldwide accompanied with a significant escalation in the prevalence of ESBL-producing Enterobacteriaceae [
5]. Unfortunately, ESBL-producing bacteria in children have come to the forefront of emerging antibiotic-resistant bacteria worldwide [
6]. β-lactamases are divided into four functional groups: penicillinases, ESBLs, carbapenemases, and AmpC-type cephalosporinases [
7]. Specifically, ESBLs is a group of plasmid-encoded enzymes that confer resistance to third generation cephalosporins [
8‐
10]. ESBLs are divided into three groups according the encoding of TEM, SHV and CTX-M genes [
11,
12]. CTX-M enzymes are the most common and are further classified into five major phylogenetic groups based on gene sequences namely, CTX-M -1, CTX-M-2, CTX-M-8, CTX-M-9, and CTX-M-25 [
11]. ESBL-producing
Escherichia coli (E. coli) and
Klebsiella pneumoniae (
K. pneumoniae) are the predominant organisms in childhood infections, and they pose significant threat to human health [
13]. These organisms are listed among the pathogens for which there are few potentially effective drugs [
1]. About 57% of bloodstream infections are caused by ESBL-producing Enterobacteriaceae, which are more likely to result in death compared to the infections caused by a non ESBL-producing strains [
14].
Urinary Tract Infections (UTIs) continue to be one of the most common cause of illness in young children worldwide [
9]. It distresses the child, concern the parent, and may cause permanent renal sequelae. The β-lactam antibiotics have traditionally been the main antibiotics for treatment of infections caused by Enterobacteriaceae, but the emergence of ESBL has rendered this class of antibiotics largely ineffective. It is therefore important to run epidemiological studies to define the epidemiology and profiles of these resistant bacteria for its impact on the development and implementation of stewardship programs.
High prevalence of extended-spectrum-β-lactamase (ESBL) and carbapenemase producing gram negative bacteria (GNB) has been reported in the Arabian Peninsula [
15]. Nonetheless, little is known about the prevalence and profile of these bacteria in pediatric population [
16] in the region. Prevalence of ESBL in urine among pediatric population has been gradually increasing in Qatar (HMC annual antibiogram). For example, the percentage of
E.coli ESBL producers have gradually increased from 18% in 2010 to 24% in 2014, and reached 31.7% in 2017 (January–June) as reported in this study.
This is the first study that describes at the molecular level the genotypic profile of ESBL producing bacteria among children with UTIs in Qatar. Our data indicates rapid increase in ESBL resistance among Enterobacteriaceae in pediatrics with UTI, which mandates rapid regulatory and monitoring reforms at the State level.
Methods
Clinical isolates and controls strains
Ethical approval for this study was obtained from the Medical Research Centre (MRC), Hamad Medical Corporation (HMC), Doha, Qatar, protocol no. 16434/16. A total of 727 urine Samples were collected between February and June of 2107 from children (0–15 years of age) hospitalized with lower UTI at the Pediatric Emergency Center-HMC. All urine analysis were performed on patients presented with symptoms, mainly fever and dysuria. Urinary catheter was applied for all patients less than or equal 2 years of age, cerebral palsy (CP) patients and patients under intermittent catheterization. Otherwise, urine was obtained from mid-stream catch. Samples that did not yield significant bacterial growth, those that had multiple organisms and samples with suspected contamination as per lab report, were excluded from the study, and no duplicate samples were collected. All of the reported cases had UTI as their primary diagnosis. For each patient, demographic data such as age, nationality, and gender were collected. Out of the above samples, 635 (87.3%) were positive for Enterobacteriaceae species which were then isolated using readymade Cystine Lactose Electrolyte-Deficient media (IMES, Doha, Qatar). Isolated bacteria were identified by MALDI-TOF (Bruker Daltonik GmbH, Leipzig, Germany) and initial antimicrobial susceptibility testing was performed by Phoenix using the NMIC/ID-5 panel (BD Biosciences, Heidelberg, Germany) according to the manufacturer’s recommendations. Both automated tests were performed at Hamad General Hospital Microbiology laboratory. All intermediate resistant isolates were considered as susceptible. Initial testing with Phoenix revealed 201 (31.7%) isolates as Extended Spectrum β -Lactamases (ESBL) producer Enterobacteriaceae. Susceptibility testing was done for 16 clinically relevant antibiotics. 110 of these samples were randomly selected for further genotypic analysis.
Standard strains, E. coli ATCC® 25,922 and E. coli ATCC® 35,218, were used as controls for antimicrobial drug susceptibility testing. E. coli NCTC® 13,461™, E. coli NCTC® 13,462™, E. coli NCTC ®13,463™, Enterobacter cloacae NCTC ®13,464™ and K. pneumonia NCTC® 13,465™, E. coli ATCC® 35,218™ and E. coli NCTC ®13,368™ were used as positive controls, for CTX-M G1, CTX-M G 2, CTX-M G 8, CTX-M G 9, CTX-M G25 25, bla TEM and bla SHV, polymerase chain reaction (PCR) assays, respectively.
ESBL phenotype confirmation
Isolates that were tested positive for ESBL by Phoenix were consequently confirmed by Double Disc Synergy Test (DDST) as previously described [
17,
18]. Briefly, synergy was determined between a 20/10 μg disc of amoxicillin-clavulanate (BD- Sensi Disc™) and 30-μg disc of ceftazidime and ceftriaxone (BD- Sensi Disc™), placed onto Mueller–Hinton agar (Oxoid Ltd., Basingstoke, Hampshire, England) inoculated with a microbial suspension of 0.5 McFarland turbidity, at a distance of 15 mm apart from the edge of the amoxicillin-clavulanate disc. The cefoxitin (30 μg, BD- Sensi Disc™) disc was placed in any available space remaining on the plate. Extension of the edge of the exhibition zone by > 5 mm towards the disc of amoxicillin-clavulanate disc, together with susceptibility to cefoxitin was interpreted as positive for the ESBL production [
18].
Molecular genotyping of ESBL genes
DNA was extracted from bacterial cultures using QIAamp® UCP pathogen mini Kit (Qiagen, Germany) following manufacturer’s instructions. Extracted DNA was then used to run PCR for seven genes and using previously published primers [
19,
20] . The conditions used for
blaTEM and
blaSHV reactions were as follows: PCR mixture was made in volume of 20 μl containing 0.5 μM of each primer, 50 ng DNA, 1× master mix (Hot star
Taq plus master mix (Qiagene, Germany)) and DPEC H2O up to 20 μl. The reaction was amplified in GeneAmp* PCR system 9700 thermocycler under the following conditions: 1. Initial denaturation at 96 °C for 5 min.; 2. 32 cycles consisting of denaturation at 96 °C for 30 s., annealing at 44 °C (
bla TEM) and at 58 °C (
bla SHV) for 45 s, and extension for 60 s. at 72 °C; and 3. A final extension cycles at 72 °C for 10 min. Multiplex PCR (MPCR) was performed in a final volume of 30 μl containing 0.23 μM of each primer (
bla CTX-M-G
(1,2,8,9 &25)), 50 ng DNA, 1× master mix (Hot star
Taq plus master mix (Qiagene, Germany) [
19]) and DPEC H2O up to 30 μl to screen for
bla CTX-M- G (1,2,8,9 & 25) genes. Amplified products were subjected to electrophoresis in 1.2% agarose (Agarose- LE, Ambion®, USA), stained with ethidium bromide (Promega, Madison, USA) and visualized using Bio-Rad gel doc system (Bio - rad, Gel Doc
tm XR System 170–8170, Canada).
Clustering of ESBL-positive isolates
An agglomerative hierarchical algorithm was used to derive a cluster analysis dendrogram to establish the relationship between individual
E. coli (
n = 95) and
Klebsiella pneumoniae (
n = 13) isolates based on the presence and absence of 7 genes (
blaTEM,
blaSHV,
bla CTXM -G 1,
bla CTXM -G2,
bla CTXM -G8,
bla CTXM -G9 &
bla CTXM –G25), whichare reported in the literature to encode for ESBL. The scores ‘1’ and ‘0’ were given for the presence and absence of bands respectively [
21‐
23]. The data obtained by scoring of the genetic profiles of different ESBL genes were subjected to cluster analysis, and hierarchical cluster dendrogram was created using Past software version 1.91 [
24]. A similarity matrix values were used for cluster analysis.
Data analysis
Data were introduced into Microsoft Excel 2010 (Microsoft Corporation, New York, USA) to generate figures and run initial analysis and further statistical analysis were done using SPSS statistics 24 (Statistical Package for the Social Science; SPSS Inc., Chicago, IL, USA). Relation between gene type and resistance of each antibiotic was cross-tabulated using non-parametric Kappa statistics; on the other hand, relation between gene type, nationality and age grouping was calculated using Pearson chi-square test. Probability value (
P value) less than 0.05 was considered statistically significant. Past software, version 1.91, was used to construct hierarchical clustering dendrogram and Jacquard’s coefficient was applied to generate the similarity values for generation of the cluster analysis [
24].
Discussion
Out of a 635 Enterobacteriaceae isolated from urine samples obtained from a cohort of children with UTI, 201 (31.7%) were found to be ESBL.
E. coli species was the most prevalent representing 83% (
n = 166) of the isolates, followed by
K. pneumoniae which represented 11% (
n = 22). This is in agreement with other studies including a recent one from Sri Lanka [
25] in which
E. coli and
Klebsiella species represented 86.8 and 13.1% of UTI infections among adult patients. Expectedly, ESBL was more predominant among females than males with a ratio of 4.9 to 1. It has been frequently reported thatUTIs occur far more frequently in girls than in boys during the first few months of life, presumably due to the shorter length of the female urethra [
26,
27]. The most affected group of our study were the children ranging in age from 0 to 5 years (70.7%), wherecolonization with
E. coli and enterococci is known to diminish after 5 years of age [
26].
Molecular genotyping of ESBL-positive isolates showed that the highest resistance was due to the presence of
bla CTX-M genes (59%), particularly
bla CTX-MG1 (89.2%).
bla TEM and
bla SHV genes contributed to only 2.7 and 0.9% of ESBL resistance, which is consistent with previous reports from the region and around the globe [
15,
28‐
31]. The high occurrence of
bla CTX-MG1 gene among pediatric population suggests high dynamic transmission-ability of the plasmid-carried gene, that contributes to the dissemination of CTX-M enzymes to the community and could cause extra horizontal transmission in healthcare facilities [
32]. In busy health care facilities, especially in developing countries, the rapid turnover of patients to accommodate multiple patients’ treatment could lead to breaching of infection control measures, such as compliance with hand hygiene, which could lead to this horizontal transmission of resistant bacteria from one patient to another. Unfortunately, we could not compare our findings to others from the region due to the scarcity of published data. However, it has been globally reported that the predominant genotype of ESBL-producing
E. coli and
K. pneumonia has changed from TEM and/or SHV to CTX-M-1 since 2006 [
33,
34], which seems to be the trend in Qatar. Interestingly, our results indicated that three
bla genes (
blaSHV,
blaTEM and
blaCTX-MG1/
blaCTX-MG9) were concurrently detected in 77% of
K. pneoumoniae isolates, whereas only 6.3% of the
E. coli isolates had combinations of the three genes (Fig.
3). On the other hand, two out of seven
bla genes were simultaneously detected in 18.2 and 3.2% of
E. coli and
K. pneumonia, respectively. These results indicate different genotypic profile of ESBL resistance between
E. coli and
K. pneumoniae in urine samples which might indicate a high transfer of genes among
K. pneumoniae than
E. coli. Concerning the cephalosporins, all ESBL-producing isolates showed 100% resistance to ampicillin, cephalothin, cefazolin, ceftriaxone and 97.3% resistance to cefepime. This finding coincided with a study carried in intensive care unit (ICU) among adult patients at HMC, Qatar, on respiratory tract and blood samples [
35]. No resistance in ESBL-producing isolates were observed against amikacin, ertapenem, meropenem and imipenem. Among the β-lactam/β-lactamase inhibitor combinations, 10% were resistant to piperacillin/tazobactam, whereas 99.1% were resistant to amoxicillin/clavulanic acid. The clinical isolates presented 24.5% resistance to gentamicin compared with 100% susceptibility to amikacin among aminoglycosides. Similarly, these findings concur with the findings from the aforementioned study in ICU patients, [
35] with an exception of piperacillin/tazobactam resistance which was reported at 22% in ICU adult patients compared to 10% in pediatrics with UTI. Resistance to piperacillin/tazobactam and ciprofloxacin were significantly lower in our study (10 and 36.4% respectively) compared with another one from Kingdom Saud Arabia (KSA) in a hospital at Dammam-Eastern province, where resistance reached63.2 and 84.6%, respectively [
30]. Authors attributed high resistance to these antibiotics due to the misuse of ciprofloxacin and restriction of aminoglycosides among adult in Dammam hospital. Whereas, ciprofloxacin is restricted for use among adult patients at HMC and very rarely used in pediatric population (personal communication, 2017).
The resistance to oral antibiotics namely, nitrofurantoin, trimethoprim/sulfamethoxazole and ciprofloxacin was found to be 9.1, 60.9 and 36.4%, respectively. Obtained results were comparable to those observed in the Middle East region [
27], which indicate rapid dissemination of multidrug resistance bacteria in the communities [
15,
36]. Worldwide, previous studies have reported higher prevalence of ESBL resistance in Europe compared to United States (US), but lower than South America and Asia [
37]. Our results indicate that the patterns of ESBL resistance in Qatar, at least in pediatric population, is more similar to Asia. It is worth noting that 80% of population in Qatar is expatriates; arriving mostly from countries in South East Asia. This could partially explain our findings. Cumulatively, our findings mandates the establishment of the antimicrobial stewardship program and formulation of guidelines for empirical use of prescription of antibiotics for UTI infections among pediatric population.
We found that the majority of ESBL (156, 77.6%) producers were multidrug resistant (resistant to 3 or more different classes of antibiotics), with the most common resistance pattern being to amoxicillin-clavulanate, ampicillin, ciprofloxacin, Sulfamethoxazole/trimethoprim in addition to cephalosporines. Plasmids encoding ESBL resistance often carry multiple genes conferring multiple resistance, thus simultaneous resistance to fluoroquinolones, aminoglycosides, tetracyclines, and trimethoprim sulfamethoxazole [
12]. Our study revealed a significant association (
p < 0.05) between the presence of TEM, nitrofurantoin and trimethoprim/sulfamethoxazole; SHV and nitrofurantoin; CTXM-G2 and piperacillin/tazobactam resistance. Thus, in our study, the use of extended-spectrum cephalosporins, ciprofloxacin, aminoglycosides (gentamicin) and trimethoprim/sulfamethoxazole is compromised, leaving carbapenems as the therapeutic option for severe infections caused by ESBL producers. In agreement with our findings, [
38] found that plasmid-mediated quinolone resistance has been associated with
bla CTX-M genes, where genes conferring resistance to aminoglycosides and tetracycline and other
bla genes have been found on the same plasmids as the
bla CTXM [
39]. Isolates in this study were all from pediatric patients in the community who attended emergency setting This highlights the therapeutic challenges posed by ESBL producers, particularly in the UTI treatment of community-onset [
40].
Using seven genes to run agglomerative hierarchical algorithm analysis [
21‐
24] our study revealed eleven clonal clusters among the 94 tested
E. coli isolates. Clonally related strains of cluster A2 were responsible for the predominant UTI in pediatrics population 67.1% (58/94), which produce only CTXM-G1 type enzyme, followed by A5 (20.2%), which produce a combination of TEM and CTXM-G1type enzymes. Five clonal clusters were detected among
K. pneumoniae with the main cluster B2 producing concurrently a combination of three enzyme types SHV, TEM and CTXM-G1. Whole plasmid sequencing for representative isolates from each cluster can be very helpful to better understand the relatedness and typing differences between the clusters and elucidating further information on the mechanism of resistance in these isolates. Unfortunately, similar molecular analysis of ESBL bacteria are lacking in the surrounding country. This limit our understanding of the trends and distribution of the strains, noting that more than 80% of Qatari population are expats that arrive from MENA region and South East Asia.