Phenotypic and Genotypic Characterization of Extended-Spectrum Beta-Lactamases Produced by Escherichia coli Colonizing Pregnant Women.

Introduction Infections caused by extended spectrum beta lactamase (ESBL) producing bacteria continue to be a challenge for choosing the appropriate therapy since they may exhibit coresistance to many other classes of antibiotics. The aim of the study was to screen pregnant women for ESBL producing bacteria in Beirut, Lebanon, to examine their phenotypic and genotypic characterization and to study the association between ESBL colonization with adverse neonatal outcomes. Method In this cross-sectional study, vaginal samples from 308 pregnant women at 35–37 weeks of gestation were studied during a one-year period. The samples were plated on MacConkey agar and selective MacConkey agar supplemented with ceftazidime. Phenotypic confirmation of ESBL production was performed by double-disc synergy test and all isolates were screened by PCR for the resistance genes blaSHV, blaTEM, and blaCTX-M. Clonal relatedness of Escherichia coli isolates was investigated by pulsed-field gel electrophoresis. Results In total, 59 women out of 308 (19.1%) were colonized by ESBL producing gram negative bacteria. Two babies born to mothers colonized with ESBL were diagnosed with sepsis. The susceptibility rates of isolates to other antibiotics were 39% to co-trimoxazole, 49.2% to ciprofloxacin, 91.5% to gentamicin, 18.6% to aztreonam and 35.6% to cefepime. Most of isolates were highly sensitive to meropenem and imipenem, with a susceptibility of 93.2%. PCR was performed on all E. coli isolates to detect the most common ESBL producing genes; blaCTX-M was the predominant gene (90.7%), followed by blaTEM (88.4%) and finally blaSHV (44.2%). PFGE analysis of 34 E. coli isolates revealed 22 distinct clusters showing more than 85% similarity. Conclusion In conclusion, this study showed that Lebanon has a high prevalence of ESBL carriage in pregnant women. Further studies that include a continuous screening of pregnant women and follow up of their newborn clinical status should be conducted to foresee the risk of transmission.


Introduction
Neonatal sepsis is a blood infection that occurs during the first month a er birth. Early onset neonatal sepsis (EOS) is considered the main cause of mortality and morbidity in neonates [1]. It occurs in the first three days of life and usually transmitted from mother to baby during delivery [1]. e infection may be also transmitted vertically when the amniotic membrane ruptures or prior to the onset of labor causing intra amniotic infection [1,2]. e organisms most frequently associated with Hindawi Infectious Diseases in Obstetrics and Gynecology Volume 2020, Article ID 4190306, 7 pages https://doi.org /10.1155/2020/4190306 Infectious Diseases in Obstetrics and Gynecology 2 EOS are Streptococcus agalactiae and Escherichia coli [1,2]. e latter accounts for about 24% of all EOS; 81% of cases occurring in preterm infants [1]. High rate of colonized extended spectrum beta lactamase producing Enterobacteriaceae in the maternal vaginal canal, in particularly Escherichia coli and Klebsiella pneumonia, have been also detected in infected newborns [3,4]. erefore, transmission to newborn may occur intrapartum and EOS secondary to E. coli may cause bacteremia with or without meningitis at the time of delivery [1].
Infections caused by ESBL producing bacteria continue to be a challenge for choosing the appropriate therapy since they may exhibit coresistance to many other classes of antibiotics [5,6]. Sparse information exists in the literature regarding ESBL producing bacteria colonization in pregnant women. It is therefore important to know the prevalence of these microorganisms in specific geographic location and formulate an appropriate screening policy. e aim of the study was to screen pregnant women for ESBL producing bacteria in Beirut, Lebanon and to examine their phenotypic and genotypic characterization. e study also aimed to explore the association between ESBL colonization with adverse neonatal outcomes.

Study Population.
A cross-sectional descriptive study was conducted from March 2016 to March 2017 involving 308 pregnant women at 35-37 weeks of gestation who were examined during antenatal checkup at different obstetrics and gynecology clinics in and around Beirut. Women signed a consent form to approve their participation in the study. One vaginal swab was collected by the attending physician from each patient attending the clinic for antenatal care. e samples were stored in Stuart medium (Oxoid, UK) at room temperature until transported to clinical diagnostic laboratory.

Data Collection.
Socio-demographic data, clinical status, and gestational history of 165 (55%) patients were collected through a questionnaire by the gynecologists. e questionnaire was developed to measure women age, education, health and delivery-related variables (delivery type and delivery time, gestational diabetes, anemia, previous miscarriage, urinary tract infection, induced labor, and contact with animals as independent variables. e questionnaire also measured neonatal outcomes (neonatal weight, height and Apgar score) as dependent variables. e questionnaire was filled by the physician who examined participants. Anonymity and confidentiality were guaranteed and a written informed consent was signed by the participants. is study was approved by the institution review board of Beirut Arab University.

Bacterial Isolation and Identification.
e clinical samples were plated on MacConkey agar and selective MacConkey agar supplemented with 1 mg/L of ce azidime (CAZ). Plates were incubated at 37°C under aerobic conditions and examined a er 24 and 48 h incubation. All isolates were identified by their culture characteristics, standard biochemical tests, and confirmed by API 20E (Biomerieux, Mary l'Etoile, France) according to the manufacturer's instructions.

Nucleic Acid Extraction and Amplification of Beta
Lactamase Genes. DNA was extracted from isolated ESBL producing E. coli strains following an overnight growth, using QIAamp DNA Mini Kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer's instructions. All isolates were screened for the resistance genes bla SHV , bla TEM , bla CTX-M by PCR using universal primers (Table 1). PCR amplification reactions were performed in a total volume of 20 l containing 2 l of 10 X PCR buffer, 0.5 l of each forward and reverse primer (10 mM), 1.2 l of MgCl 2 (25 mM), 2 l of dNTP mix (20 mM), 0.5 of AmpliTaq Gold DNA polymerase ( ermo Fisher Scientific, Waltham, Massachusetts) and 2 l of DNA template. e amplification cycles were as follows: an initial denaturation at 95°C for 15 min; followed by 40 cycles of 95°C for 30 s, 56°C (bla SHV , bla TEM ) or 58°C (bla CTX-M ) for 30 s, and 72°C for 60 s; and with a final extension at 72°C for 10 min. e amplified PCR products were subjected to electrophoresis at a 1.5% agarose gel in 1 × TAE buffer.

Pulsed-Field Gel Electrophoresis (PFGE)
. PFGE was performed using the Xba I restriction enzyme ( ermo Fisher Scientific, Waltham, MA) for E. coli isolates (푛 = 35) identified in pregnant women according to the PulseNet protocol. Clonality and genomic relatedness were determined using the CHEF MAPPER (Bio-Rad, Austin, TX). e BioNumeric fingerprinting so ware (Applied Maths, Belgium) was used to analyze the profile and generate a dendrogram describing the relationship among the isolates. PFGE patterns were analyzed using Tenover's criteria (same cluster if the dice similarity index was >85% and < than 6 bands difference).

Statistical
Analysis. Antibiotic susceptibility rates were calculated using frequency and percentages. Colonization with ESBL as independent variable was correlated with the newborn height, weight, and Apgar score (overall assessment of new born well-being used immediately following the delivery of the baby) as dependent variables, taking into consideration other possible confounding variables including mother's age, mother's education, previous miscarriage, delivery week, delivery type, induced labor, recurrent UTI, gestational diabetes, anemia, vaginal discharge, and contact with domestic animals. Beta coefficient, which measures the magnitude of effect of the independent variables on the dependent variable in a multiple regression analysis, was calculated. Statistical significance was calculated using -value and confidence intervals. e effect of colonization with ESBL on categorical outcome variables (Gestational diabetes, vaginal discharge, induced labor and recurrent UTI) was explored using the test of independence Chi-square. -values were computed considering 푝 ≤ 0.05 as significant results.
In this study, all isolates were detected by three phenotypic methods. e result of disc diffusion susceptibility testing of isolated ESBLs revealed that all were resistant to amoxicillin-clavulanic acid, ce azidime and cefotaxime. e susceptibility rates of isolates to other antibiotics were 39% to co-trimoxazole, 49.2% to ciprofloxacin, 91.5% to gentamicin, 18.6% to aztreonam and 35.6% to cefepime. Most of isolates were highly sensitive to meropenem and imipenem, with a susceptibility of 93.2%. None of isolates was sensitive to all antibiotics and all of isolates showed resistance to more than two antibiotics. e frequency of multidrug-resistant to three and more antibiotics was 25.4% of isolates. Figure 1 and Table  2 illustrate resistance pattern for different isolated ESBL species among the 59 isolates.

Prevalence of Extended Spectrum Beta Lactamases
Genes in E. coli Positive Isolates. PCR was performed on all E. coli isolates to detect the most common ESBL E. coli producing genes; bla CTX-M was the predominant E. coli gene (90.7%), followed by bla TEM (88.4%) and finally bla SHV (44.2%) ( Table  3). irty eight (88.4%) isolates carried more than one type of b-lactamase genes. Coexistence of the bla CTX-M and bla TEM was detected in 19 isolates (44.2%), bla CTX-M and bla SHV in 3 isolates (6.8%) and bla CTX-M , bla SHV and bla TEM in 16 isolates (37.2%). e carriage of a single gene, bla CTX-M or bla TEM gene was observed in one and three isolates, respectively.

Pulse Field Gel Electrophoresis
Analysis. PFGE analysis of 34 E. coli isolates revealed 22 distinct clusters showing more than 85% similarity. Cluster 4 was prominent in 8 (23%) isolates. Cluster 10, 11, 13, 15, and 19 were seen in more than one E. coli isolates. Clutser 4 isolates had different antimicrobial susceptibility profile (Figure 2). Samples 25 and 26 had very similar profiles, the difference between both isolates was the additional presence of SHV gene.

Association between the Presence of Infection and Neonatal
Outcomes. In the current study, retrospective data collection showed that two babies born to mothers colonized with ESBL were diagnosed with sepsis. However, both mother and baby isolates were not available for genetic comparison.
Results of the three multiple regression models with neonatal outcomes (weight, height and Apgar score) as dependent variables are displayed in Table 4. Results revealed that ESBL colonization had negative association with height of the newborn, although the association was not statistically significant. However, significant positive association between delivery week and newborn height and significant negative associations between the mother age, delivery type, and newborn height were noted. e height of the newborn increased 0.38 cm with one week increase in delivery time ( value = 0.003) and decreased 0.23 cm with the increase in the mothers' age ( value = 0.04). Newborn height also decreased with C-section (Beta = 0.46; value = 0.004). On the other hand, ESBL colonization had significant negative association with the weight loss of the newborn (Beta = −0.31; 푝 = 0.009 ). C-section also had significant negative association with weight (Beta = −0.41; p value = 0.009), while delivery time had significant positive association with the newborn weight; there was 0.33 g increase in weight with an additional delivery week ( value = 0.008). ESBL colonization had negative association with Apgar score, however without statistical significance. e other covariates did not yield significant associations. inappropriate prescription of antibiotics, and the extensive use of over-the-counter antibiotics aggravated the emergence of antibiotic resistance [9]. In 2013, the prevalence rate of ESBL production of E. coli and Klebsiella species reached 32.3-29.2%, respectively [10]. In 2016, the percent susceptibility of Enterobacteriaceae to third-generation cephalosporins was 59% [11].
is study is the first in Lebanon that evaluates the prevalence of ESBL in pregnant women. Our study showed a 19.1% prevalence, which is similar to previous studies that reported a prevalence that ranges from 7.5% to 25% [12][13][14]. e high rate of resistant to cefipime, trimethoprim salfamethoxazole, aztreonam, and ciprofloxacin reflects the increasing prevalence of resistance in Lebanon. Gentamycin, meropenem, and impenem were, however, susceptible to most isolated strains and can be used as intrapartum antibiotic prophylaxis for the prevention of the infection. Antibiotic resistant strains may reside in the genital track of the mother and may be transmitted to new born during delivery. Preterm infants are at high risk of ESBL producing Enterobacteriaceae sepsis in neonatal care unit. [3,4,15]. Maternal-neonatal transmission has been reported, where identical strains were identified from mother to infant through vertical transmission [16][17][18].
It is not well known whether the usage of antibiotics during the delivery may decrease the risk of acquiring ESBL in new born. ere are controversial data showing that antibiotics are able to reach the fetus by crossing the placenta, hence increasing the risk of ESBL-PE acquisition in newborns (ESBL-PE acquisition in neonates). So far, in Lebanon, there is no agreement regarding surveillance of pregnant women for ESBL colonization. screening of pregnant women for species is done only for streptococcus agalactiae as a part of antenatal checkup in Lebanon as well as in other middle east 4. Discussion e antimicrobial resistance is becoming a major threat in Lebanon. e preexisting colonization of the gastrointestinal track with antibiotic resistant organisms in Lebanese patients has been previously reported [8]. Empirical therapy,  Infectious Diseases in Obstetrics and Gynecology 6

Conclusion
In conclusion, this study showed that Lebanon has a high prevalence of ESBL carriage in pregnant women. It should be noted that the current study has few limitations; Only pregnant women were examined without conducting a follow-up study on their newborn to measure the rate of transmission. In addition, we did not screen for nonpregnant women which may have enabled the authors to identify if pregnancy could be a risk factor for acquiring resistance strains.

Data Availability
All authors confirm that all data and material are available.

Ethical Approval
Study activities were reviewed by Beirut Arab University International Review Board (IRB) under IRB# 0041-S-P-0336.

Conflicts of Interest
All authors declare that they have no conflicts of interest.

Authors' Contributions
NG was responsible for the study design, performed and analyzed the experiments in addition to data analysis and writing up the manuscript. MEC was responsible for the study design, supervised and analyzed the experiments, and was responsible for writing the manuscript. AE and RH were responsible countries [19,20]. erefore, minimal risk of transmission and infection may reside. e rate of mother-to-infant transmission for other infections should be investigated by conducting further studies that include a continuous collection and screening of mothers' vaginal samples and follow up of their newborn clinical status. In the current study, two babies were diagnosed with neonatal sepsis. In addition, it is also important to conduct long-term follow-up of children born to ESBL infected women who have been exposed to antibiotics before and a er delivery. Early diagnosis may therefore minimize the risk of transmission to the baby. It is also essential to study virulence factor such as the K1 capsule that may cause meningitis in neonates. e risk factor of acquiring ESBL resistance includes the overuse of antibiotics [21], in addition to the transmission of resistance genes from community, livestock, animals, and environment [22]. In the current study, we demonstrated the high prevalence of CTX-M and TEM genes in community patients. e results were consistent with other studies done in the region; however, most of the studies were investigating strains isolated from hospitalized individuals with the common presence of CTX-M-15 gene [23,24]. CTX-M-producing E. coli isolates are o en coresistant to various antibiotic classes, which include co-trimoxazole, the aminoglycosides, and the fluoroquinolones [25]. In our study, 18 E. coli strains carrying CTX-M gene were resistant to ciprofloxacin and 24 strains were resistant to cotrimoxazole.
PFGE analysis demonstrated a genomic diversity among E. coli strains with the presence of one major circulating clone among ESBL producer strains identified in pregnant women. Further analysis should have been done on the neonates in order to investigate if transmission of stain has occurred. for sampling and clinical interpretation. GM, ABF, and NS performed the PFGE experiment and the analysis. AG was responsible for the epidemiological and statistical analysis of the data. RD, WAF, and HY revised the manuscript. All authors reviewed and approved the manuscript.