Phenotypic detection of extended spectrum beta-lactamase in multidrug-resistant Escherichia coli from clinical isolates in Niamey , Niger

1 Laboratoire de Biologie Moléculaire d’Epidémiologie et de Surveillance des Agents Transmissibles Par Les Aliments (LaBESTA). Centre de Recherche en Sciences Biologiques, Alimentaires et Nutritionnelles (CRSBAN). Département de Biochimie-Microbiologie. UFR en Sciences de la vie et de la Terre. Ecole Doctorale Sciences et Technologies. Université Ouaga I Pr Joseph KI-ZERBO, 03 BP 7021 Ouagadougou 03, Burkina Faso. 2 Laboratoire de Biologie, Hôpital National de Niamey (HNN), BP 238 Niamey, Niger. 3 Laboratoire de Biologie, Hôpital National Lamordé (HNL), BP 10146 Niamey, Niger. 4 Laboratoire de Biotechnologies, Faculté des Sciences et Technologies Université Abdou Moumouni de Niamey B.P.: 12022, Niamey, Niger. 5 Laboratoire de Biochimie Hôpital National de Niamey (HNN) BP 238, Niamey, Niger.


INTRODUCTION
Escherichia coli is a commensal of the human gut and one of the most frequently isolated bacteria from clinical specimens (Quinet et al., 2010).It plays an important role as a member of the gut microbiota; however, pathogenic strains also exist, including various diarrheagenic E. coli pathotypes and extraintestinal pathogenic E. coli that cause illness like bacteremia, bladder infections, meningitis or pus (Dias et al., 2009;Smith et al., 2010;Fratamico et al., 2016).The discovery of antibiotics has been a humanity relief because these remedies have significantly reduced the incidence of infectious diseases, especially in developing countries (Guessennd et al., 2008).E. coli is one of the most common clinical pathogens causing nosocomial infection.For a long time, the widespread use of antibiotics to treat E. coli infectious disease has rapidly increased the multidrug resistance (MDR) of E. coli (Trecarichi et al., 2012;Kanwar et al., 2013) especially with those strains producing ESBL (Bush, 2001;Woerther et al., 2013).The appearance of ESBL stated in the 1980s and widely distributed in the world (Knothe et al., 1983;Bradford, 2001) and conferred increased resistance to beta-lactams except carbapenems and cephamycins (Patterson, 2001;Masterton et al., 2003).ESBLs are plasmid mediated and the genes encoding these enzymes are easily transferable among different bacteria (Todar, 2012).Most of these plasmids not only contain DNA encoding ESBLs but also carry genes conferring resistance to several nonβ-lactam antibiotics (Rankin and Svara, 2011).The presence of ESBL in clinical isolate has been documented as a very serious problem and a significant trait to: quick survival of patients in the hospital, high economic burden, loss of hours in life's activities and high treatment failure (CDC, 2010).The phenotypic methods are currently the gold standard in determination of susceptibility or resistance of clinical isolates.The most widely used methods to screen ESBL are E-test, or double-disk synergy test (DDST) (EUCAST, 2014).Several reports have described the prevalence of ESBLs in the Middle East North Africa region and most of the Gulf Cooperation Countries (Zowawi et al., 2013).However, there is insufficient scientific data on the prevalence of ESBLs available from the State of Niger.
This study aimed to determine the prevalence of ESBLproducing among MDR E. coli isolates from various clinical samples at "Hôpital national de Niamey" and "Hôpital national lamordé", Niger.

Study design and site
The present prospective study was conducted on routine specimens received at the bacteriology laboratory of"Hôpital National de Niamey" and the "Hôpital National Lamordé", with a

Isolation and identification of E. coli
Bacterial isolates were obtained from various clinical specimens.Stool samples were inoculated on eosin methylene blue agar (EMB, Merseyside UK), urine samples on cystine lactose electrolyte deficient agar (CLED, Liofilchem), pus and vaginal swabs were cultured on blood and chocolate agar and were incubated at 37°C for 18 to 24 h.Blood samples were inoculated and incubated in culture bottles with Bact/Alert 3D 60 Biomérieux at 37°C.The isolates were identified according to the procedures described by Cheesbrough et al. (2005) and were confirmed using gallery API 20 E system (Biomérieux, France).The clinical isolates were preserved at −70°C for further analysis.
Multidrug resistant E. coli was defined as resistance to at least three classes of antibiotics.

Detection of extended-spectrum beta-lactamase (ESBL) by DDST
All MDR E. coli isolated were screened for ESBL.The double-disk synergy test (DDST) was performed for the phenotypic detection of ESBL producers according to the CA-SFM recommendations, using ceftazidime (30 µg), aztreonam (30 µg), cefotaxime (30 µg) disks and were placed 25 mm (center to center) from the amoxicillin/clavulanic acid (20/10 µg) disk on Mueller-Hinton agar (CA-SFM, 2012).After inoculation, the plates were incubated at 37°C; the presence of a keyhole effect was recorded 24 h after incubation.The DDST was performed in parallel to the antibiogram.

Detection of extended-spectrum beta-lactamase by DDST using cloxacillin
Cloxacillin test was performed for MDR isolates naturally producing inducible céphalosporinase (AmpC).The production of ESBL was inferred by a synergy image as previously described (Drieux et al., 2008) to September 2015, all MDR E. coli that were ESBL negative for DDST, were tested by DDST using cloxacillin.
The DDST was also performed with cloxacillin (200 μg/ml) containing MH agar plates (bioMérieux).In the situation of the absence of growth of E. coli on MHA with cloxacillin at a concentration of 200 μg/ml, cloxacillin at a concentration of 100 μg/ml was used (Drieux et al., 2008).

Ethical consideration
All isolates obtained were from biological specimens on clinical routine examinations of patients, with the authorization of the laboratory director and hospital director.

Data analysis
Data were analyzed using Excel, Microsoft® Office 2013.Chi square was used to determine the statistical significance of the data.Statistical significant difference was considered with a p-value ˂ 0.05.
Those MDR E. coli (168) that were negative to ESBL were tested with DDST using cloxacillin.Of these isolates, ESBL prevalence of 4.8% (9) was observed (P = 0.0013).This indicated a significant difference between the two methods.

Prevalence of ESBL E. coli according to demographic characteristic of the studied patients
The demographic characteristic of the studied patients is summarized in Table 3.Of the 217 MDR E. coli, 150 (69.1%) were isolated from outpatients, and 67 (30.9%) from inpatients (P= 0.0002).However, ESBL prevalence of 25.3 and 28.4% from outpatients and inpatients were

Prevalence of ESBL E. coli according to biological specimens
Most of the MDR E. coli were isolated from the urine samples (67.3%) followed by stool samples (26.3%) as shown in Table 4.However, only 26.7 and 26.3% of urine and stool isolates, respectively were ESBL-producers.On the other hand, in spite of their small number, 25% of isolates from pus and blood samples were ESBL producers.

DISCUSSION
The number of infections due to ESBL E. coli is increasing, especially in African countries (Manyahi et al., 2014).In this study, the authors investigated the prevalence of ESBL production by MDR E. coli isolated from clinical samples sent to the two main hospitals of Niamey.The antimicrobial susceptibility tests showed an important level of resistance of antibiotics classes.Thus, for beta-lactams classes, resistance frequency of 93.1% was observed for amoxicillin-clavulanate, 92.6% for cefotaxim and 97.2% for ceftazidim.Similar results have been reported in Nigeria with a prevalence of 89.71% for amoxicillin-clavulanate 79.47% for cefotaxim and ceftazidim 41.03% (Odumosu and Akintimehin, 2015).On the other hand, co-resistance was shown for different antibiotics such as ofloxacin (77.4%), ciprofloxacin (84.9%) and nalidixic acid (91.2%).Similar results were observed in Ivory Coast with a prevalence of 70.2% for ciprofloxacin and 76.8% for nalixidic acid (Guessennd et al., 2008).Such level of resistance could be due to abusive prescription of antibiotics by professionals of health care without prior laboratory investigations or parallel care at home, self-medication and also the use of street drugs which is very spread in Africa (Yandai et al., 2014).Nevertheless, this study showed a susceptibility of 95.4% for imipenem, this frequency was similar to published data in Burkina Faso, which showed a susceptibility of 100% for imipenem (Sanou et al., 2015).In all the 217 MDR E. coli isolated in this study, ESBL prevalence of 26.3% was observed.Similar prevalence was found in some African countries such as in Benin (22%) (Ahoyo et al., 2007), Nigeria (20%) (Onwuezobe and Orok, 2015), Niger (30.9%) (Woerther et al., 2011) and Chad (20.09%) (Yandaï et al., 2014).As compared to previous studies, the current study results are lower than that observed in Burkina Faso (38.3%) (Dembélé et al., 2015) and Senegal (52%) (Lo et al., 2014).However, these results are higher than that observed in Tanzania (15.1%) (Mshana et al., 2016) and the Libyan community (13.4%) (Ahmed et al., 2014).This wide variation in prevalence was probably due to differences in type of samples collected.
The study showed the prevalence of ESBL producers based on gender and indicated that females had a higher prevalence rate (28.3%) than males (24.6%).No differences were apparent between ESBL-producing E. coli with gender distribution (Akanbi et al., 2013).However, contradictory observations were found in Nigeria (Yusuf et al., 2013).In this study, the majority of ESBL-producing E. coli was isolated from inpatients (28.4%) as compared to outpatients (25.3%).The rates of ESBL-producing E. coli were higher among inpatients (22.82%) than the outpatients (18.11%) as reported in Chad (Yandai et al., 2014).
Among the patients studied, the highest prevalence was observed in age group over 65 years (50%) followed by age group 45 to 65 years (38.1%) and the age group under 5 years (24.8%).Higher median age was observed among individuals colonized with ESBLs as reported from Tanzania (Mshana et al., 2016).Although, previous study in Guinea-Bissau showed that ESBL E. coli prevalence was high in all age groups, among the youngest 27% were carriers in the ages 0 to 3 months.This indicates that colonization with ESBL-producing bacteria often occurs early in life in this population (Isendahl et al., 2012).
This study showed that MDR E. coli were isolated from a variety of clinical samples.Thus, it was found that urine samples had the highest proportion of ESBL (26.7%) followed by stool samples (26.3%).However, the same prevalence (25%) was observed from blood and pus samples.The major ESBL E. coli producer (18.2%) was isolated from urine samples (Raut et al., 2015).Ouedraogo et al. (2016) found that blood cultures had the highest proportion of ESBL isolates.In previous study, the distribution of ESBL producers was most prevalent in blood (22.2%) and urine samples (17.6%).This was followed in that order by stool (15.8%), urogenital swabs (14.3%) and wound swab (13.5%) while the least prevalence was observed in ear swab specimens (3.2%) (Yusuf et al., 2013).This difference of ESBL producers among clinical source could be due to lower number of some samples.
Therefore, infection control strategies, with the rational use of antibiotics could be an important factors to reduce the spread of ESBL.Further investigations, including molecular characterization of different ESBL was necessary to understand the spread of resistant bacteria.

Conclusion
This study demonstrated the high occurrence of ESBL produced by E. coli isolated in two hospitals of Niamey.All the age groups were concerned with high resistances of the ESBL-producing isolates to antibiotics classes.Urine and stool samples had a higher prevalence of ESBL producers in Niamey.Hence, it was necessary to develop a nosocomial infection control and antimicrobial surveillance system in all health centers in order to avoid emergence and clonal spread of ESBL E. coli.
. From July 2014 to September 2014 and then from July 2015 *Corresponding author: E-mail: juniorfodym@gmail.com.Tel: +22796577781 or +22667231547.Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution License 4.0 International License

Table 2 .
Prevalence of ESBL among MDR E. coli.

Table 3 .
Prevalence of ESBL E. coli according to demographic characteristic of the studied patients.

Table 4 .
Prevalence of ESBL E. coli according to biological specimens.