High prevalence and characterization of extended-spectrum ß-lactamase producing Enterobacteriaceae in Chadian hospitals

Background Extended-spectrum ß-lactamase-producing Enterobacteriaceae (ESBL-PE) represent a major problem in the management of nosocomial infections. However, ESBL-PE are not systematically monitored in African countries. The aim of this study was to determine ESBL-PE prevalence in patients from three hospitals in N’Djamena, the capital city of Chad, and to characterize the genetic origin of the observed resistance. Methods From January to March 2017, 313 non-duplicate isolates were recovered from various clinical specimens obtained from 1713 patients in the three main hospitals of N’Djamena. Bacterial species were identified by matrix-assisted laser desorption ionization-time of flight mass spectrometry. Susceptibility to 28 antibiotics was tested using the disk diffusion method on Müller-Hinton agar, and ESBL production was confirmed with the double-disc synergy test. The most prevalent ESBL genes associated with the observed resistance were detected using multiplex PCR followed by double-stranded DNA sequencing. Results Among the 313 isolates, 197 belonged to the Enterobacteriaceae family. The overall ESBL-PE prevalence was 47.72% (n = 94/197), with a higher rate among inpatients compared with outpatients (54.13% vs. 34.37%). ESBL-PE prevalence was highest in older patients (≥60 years of age). E. coli was the most common ESBL-producer organism (63.8%), followed by K. pneumoniae (21.2%). ESBL-PE were mainly found in urine samples (75%). The CTX-M-1 group was dominant (96.7% of the 94 ESBL-PE isolates, CTX-M-15 enzyme), followed by the CTX-M-9 group (4.1%). 86% of resistant isolates harbored more than one ESBL-encoding gene. ESBL production was also associated with the highest levels of resistance to non-β-lactam drugs. Conclusions The prevalence of ESBL-PE harboring resistant genes encoding ESBLs of the CTX-M-1 group was high (48%) among clinical isolates of three main hospitals in Chad, suggesting an alarming spread of ESBL-PE among patients. Electronic supplementary material The online version of this article (10.1186/s12879-019-3838-1) contains supplementary material, which is available to authorized users.


Background
Extended-spectrum ß-lactamase-producing Enterobacteriaceae (ESBL-PE) represent a major problem in the management of nosocomial infections, resulting in prolonged hospital stays, increased hospital charges, and higher mortality and morbidity rates [1]. ESBLs confer resistance to many antibiotics, such as penicillins, cephalosporins and aztreonam, but not to cephamycins, moxalactam and carbapenems. Klebsiella pneumoniae and Escherichia coli are the main ESBL-producing organisms worldwide. Although at a lower frequency, these enzymes have also been detected in several other members of the Enterobacteriaceae family, such as Enterobacter spp., Citrobacter spp., Proteus spp. and Morganella morganii. [2][3][4]. Therefore, all these species can contribute to ESBL spread in hospital settings. Moreover, due to the coexistence of various modifying enzymes on the same plasmid, ESBL-PE often are resistant also to fluoroquinolones, aminoglycosides, trimethoprim sulfamethoxazole and tetracycline. Thus, ESBL-PE frequently display a multidrug resistance phenotype and are an important cause of treatment failure [5,6].
ESBLs are encoded by different genes [7] inserted in genetic mobile elements, such as plasmids, that facilitate their spread between bacterial species. The most common ESBLs belong to the CTX-M, SHV and TEM families [8,9]. The CTX-M family, particularly CTX-M-15, has emerged worldwide, and is now the most common ESBL type in hospitals and in the community [10]. Although ESBL-mediated bacterial resistance is recognized as an important health problem, limited data are currently available on ESBL-PE prevalence and molecular characterization in Sub-Saharan Africa. Particularly, to our knowledge, there is no study on ESBL-PE prevalence in clinical isolates in Chad.
The purpose of this study was to determine the prevalence and genetic characteristics of ESBL-PE in three main hospitals in Chad.

Setting
This study was conducted in the three main hospitals of Chad from January to March 2017. These three hospitals are located in N'Djamena, the capital city of Chad (1.5 million inhabitants) and are: (i) the National Reference General Hospital (HGRN), a university teaching hospital and one of the first national reference health facility. This hospital has 750 beds, with 8517 admissions and 50,896 outpatients in 2016; (ii) the Mother and Child Hospital (HME), a university teaching hospital and the reference mother-child hospital in Chad. It has a capacity of 261 beds (including an intensive care unit), with about 5000 admissions and 45,000 outpatients in 2016; and (iii) the Renaissance Hospital (HR), a tertiary healthcare facility designed to receive patients with complicated/chronic diseases from other healthcare centers. It has 250 beds and 8 intensive care unit beds. In 2016, 1457 inpatients were admitted among 23,909 consultations.

Sample collection and identification
We analyzed 1713 consecutive clinical specimens (urine, surgical wound, pus, stool, sperm and blood samples) sent to the microbiology laboratory of each of these three hospitals (HME: n = 623, HGRN: n = 505, HR: n = 585). From these specimens, 313 non-duplicated and clinically significant bacterial isolates were obtained. Identification of the bacterial species was performed using biochemical tests and then confirmed by matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry (Bruker Daltonics, Bremen, Germany).

Molecular characterization of ESBL and associated genes
DNA was extracted from one single colony of each isolate by incubation in a final volume of 100 μL of distilled water at 95°C for 10 min followed by centrifugation. The presence of the bla CTX-M (CTX-M group 1, 2, 8, 9 and 25), bla TEM , bla SHV and blaOXA-like genes was assessed using a multiplex PCR method following the protocol by Dallenne et al. 2012 [12]. Primers are listed in Table 1. The cycling conditions were: 95°C for 10 min, followed by 30 cycles of denaturation at 95°C for 40s, annealing at 55°C for 40s, elongation at 72°C for 1 min, and a final elongation step at 72°C for 7 min. DNA samples from reference bla CTX-M , bla TEM , bla SHV and bla OXA-like -positive strains were used as positive controls. The plasmid-mediated quinolone resistance (PMQR) gene (qnr (A, B, C, D, S), aac (6′)-Ib-cr, qepA and oqxAB) and the aminoglycoside resistance-conferring 16S rRNA methylase genes (armA, rmtB and rmtC) were assessed using PCRs as previously described [13,14]. PCR products were visualized after electrophoresis (100 V for 90 min) on 2% agarose gels containing ethidium bromide. A 100 bp DNA ladder (Promega, USA) was used as marker size. PCR products were sequenced bidirectionally on a 3100 ABI Prism Genetic Analyzer (Applied Biosystems). The sequencing data were analyzed online using the BLAST tool available at the National Center for Biotechnology Information web page (https://blast.ncbi.nlm.nih.gov/Blast.cgi).

Statistics
Statistical analyses were performed using the Epi Info software, version 3.5.3 (Centers for Disease Control and Prevention, Atlanta, GA, USA). Differences in the proportion of ESBL-producers between patient groups were assessed using the Chi-square test, while associations between the presence of ESBL-encoding genes and categorical variables (sex, age and source of infection) were tested using multinomial logistic regressions. A p value < 0.05 was considered as statistically significant.

Resistance patterns in ESBL-producing and non-ESBLproducing Enterobacteriaceae
Resistance to antibiotics that are not hydrolyzed by ESBLs was more frequent in ESBL-PE than in non-ESBL-PE isolates, except for fosfomycin (Fig. 1)

Characterization of ESBL-encoding genes
The results of the PCR and sequencing analyses showed that the CTX-M group 1 was the most common (96.4% of isolates) ESBL type, and all the CTX-M-positive isolates carried the bla CTX-M-15 gene (Table 4). CTX-M group 9 was present in 4.1% of all ESBL-PEs (bla CTX-M-27 and bla CTX-M-14 ). The bla CTX-M-27 gene was detected only in E. coli isolates, and the bla CTX-M-14 gene only in K. pneumoniae isolates. The bla SHV , bla CTX-M group 2, 8 and 25 genes were not detected in any of the ESBL-PE isolates. 86% of isolates carried more than one were associated with one to two other ß-lactamase genes (bla TEM-1 and bla OXA-1 ), 62% were associated with one to three PMQR (qnrB, qnrD, qnrS, aac(6′)-ib-cr, oqxAB and qepA) and 9% carried one to two 16S rRNA methylase genes (armA, rmtB and rmtC) Table 4. Two E. coli isolates harbored the bla CTX-M-15 gene in association with bla CTX-M-27 , while one K. pneumoniae isolate carried the bla CTX-M-14 gene alone.

Discussion
The study reveals an ESBL-PE prevalence of 48% among clinical isolates in three major Chadian hospitals. Our results also confirm the spread of CTX-M-15 genes in isolates from African patients, and the finding that ESBL-PE display co-resistance to other antibiotic classes.
ESBL-PE prevalence varies widely between geographic areas. Low prevalence rates have been reported in Europe, USA and North America [15,16], while high rates are usually observed in South America, Asia [17] and some African countries [18]. In Sub-Saharan Africa, and particularly in Central Africa, limited data are available on ESBL-PE. The prevalence found in our study (48%) is similar to the one reported for other African countries, such as Ghana (49.4%) [19], Gabon (50%) [20], Burkina Faso (58%) [21] and Cameroon 55.3% [22], and higher than in Nigeria (20.9%) [23] and Central African Republic (19.3%) [24]. Therefore, it confirms the spread of these bacteria in the African continent. A possible explanation for such high ESBL-PE prevalence is the high selective pressure generated by an important use of  OXA-1 (2) n number, PMQR plasmid-mediated quinolone resistance beta-lactam antibiotics in African countries, where they are frequently proposed as first-line treatment for bacterial infections caused by Enterobacteriaceae [25]. Other factors that contribute to their spread include non-prescription antimicrobial use, self-medication, poor hygiene, high burden of infectious diseases, consumption of counterfeit drugs, lack of antimicrobial resistance detection systems, and absence of diagnostic tools [26][27][28]. E. coli and K. pneumoniae were the most common ESBL-PE isolates and most of these isolates were from urine samples, in agreement with previous findings in India [29]. Urinary Tract Infection (UTI) is the most frequent bacterial infection worldwide in patients with nosocomial and community-acquired infections, and Enterobacteriaceae (mainly E. coli and K. pneumoniae) are generally the causal agent [30,31].
ESBL-PE prevalence was significantly higher in isolates from inpatient than outpatient, as previously reported in Ghana and Rwanda [19,31]. This pattern could be explained by the extensive use of ceftriaxone and cefotaxime as empirical antibiotic treatment in Chadian hospitals. Moreover, hospitalization has been identified as a high-risk factor for ESBL-PE infection, because ESBL-encoding genes are carried via plasmids that can be easily disseminated among the different bacteria that contaminate hospitalized patients [26,32]. Both factors could be operating at the same time, and further research is needed to determine their contribution to the observed pattern. Like in previous studies, ESBL-PE were more frequent (p = 0.002) in isolates from older patients (≥60 years of age) [33]. This could be explained by the frequent administration of antibiotic therapy to older patient.
Regarding the association of resistance to different antibiotic classes, this study shows a positive association between ESBL-PE and resistance to quinolones, aminoglycosides (except amikacin), tetracycline, chloramphenicol and co-trimoxazole (trimethoprim/sulfamethoxazole), as previously reported in Burkina Faso and Gabon [21,34]. The resistance to other antibiotic classes in ESBL-PE isolates is alarming, because it could further restrict the choice of adequate empirical therapy for the treatment of infections caused by these bacteria. In our study, isolates were susceptible to imipenem, ertapenem and amikacin. However, these drugs should be used with caution for empirical treatments in order to avoid emergence of carbapenem-resistant Enterobacteriaceae.
The interpretation of the findings of this study is limited by the fact that there was no knowledge of the patients' previous antibiotic treatments. Indeed, antibiotic treatments prior to sample collection could have favored the transient selection of resistant bacteria, and thus increased ESBL-PE prevalence compared with patients who were not previously treated with antibiotics.

Conclusions
This report reveals a high ESBL-PE prevalence (48%) and the predominance of the CTX-M-15 enzyme among clinical isolates in three major Chadian hospitals. This emphasizes the urgent need to rationalize the use of antibiotics in hospital settings and to implement a national surveillance system for antibiotic-resistant bacteria to develop empirical treatment guidelines.
We also recommend further investigations to monitor carbapenem resistance and to determine whether healthy individuals act as ESBL-PE reservoirs in the community. These studies will contribute to better understand the mechanisms responsible for the spread of ESBL-PE in hospitals and communities.

Additional file
Additional file 1: Raw data generated and analyzed during the current study. (XLSX 35 kb) Abbreviations ESBL-PE: extended-spectrum ß-lactamase-producing Enterobacteriaceae; HGRN: National Reference General Hospital; HME: Mother and Child Hospital; HR: Renaissance Hospital; MALDI-TOF: matrix-assisted laser desorption ionization-time of flight; OR: Odd ratio

Acknowledgments
We would like to thank IRD and CHU. We thank Elisabetta Andermarcher for assistance in preparing and editing the manuscript.

Funding
This study was supported by a grant from Islamic Development Bank (IDB). The IDB were not involved in the study design, collection, analysis, interpretation of data and in writing the manuscript.

Availability of data and materials
All data generated or analysed during this study are included in this published article and its Additional file 1. Ethics approval and consent to participate This study was approved by the ethics board of each hospital and by the Chadian Ministry of Public Health (N o 676/PR/PM/MSP/SE/SG/DRGP/DRH/ SGF/16). Informed written consent was obtained from all subjects and from at least one parent for each child before enrolment in the study.