Patterns of Antimicrobial Resistance and ESBL Carriage in Enterobacteriaceae Isolates from Broiler Chicken in the West Region of Cameroon: A Cross-sectional Study

Background: The emergence of multidrug-resistant foodborne pathogens of animal origin is a growing concern. In particular, antibiotic resistance in Enterobacteriaceae of clinical importance has been on the rise. Identifying and monitoring resistance patterns in residual intestinal microora in poultry are of great signicance in the containment of antimicrobial resistance. The current study aimed to detect Enterobacteriaceae among broiler chicken and determine key antibiotic resistance patterns in isolates from poultry chicken in the West Region of Cameroon. Results: 275 cloacal swabs were collected from 28 poultry farms in 11 locations in 5 out of the 8 Divisions in the West Region. All samples tested positive for Enterobacteriaceae with an average of 3 different colonies per sample. 394 isolates were obtained belonging to 12 different Genera of Enterobacteriaceae distributed as 81 (20.56 %) Escherichia spp, 74 (18.78 %) Salmonella spp, 39 (9.90 %) Klebsiella spp, 38 (9.64 %) Proteus spp, 34 (8.63 %) Citrobacter spp, 31 (7.87 %) Enterobacter spp, 28 (7.87%) Providencia spp, 19 (4.82%) Hafnia spp, 15 (3.30 %) Shigella spp, 14 (3.55 %) Raoultella spp, 13 (3.30 %) Yersinia spp and 8 (1.78 %) Morgenella spp. Antibiotic susceptibility testing on isolates showed the following overall resistance to the various antibiotics tested: amoxicillin 345 (87.8%), amoxicillin/clavulanic acid 227 (57.8%), ceftriaxone 79 (20.1%), cefotaxime 65 (16.5%), imipenem 16 (4.1%), gentamicin 58 (14.5%), amikacin 12 (3.1%), ciprooxacin 142 (37.1%), levooxacin 124 (33.1%) and doxycycline 380 (96.7). 217 (55.1 %) were resistant to at least one antibiotic A total of nine (9) rapid screening biochemical tests 1,18 were performed on the isolates including: glucose and lactose fermentation, and hydrogen sulphide (H 2 S) and gas production using Kligler Iron Agar (KIA), urease activity and indole production using urea broth and Kovacs’ reagent, mannitol fermentation and motility using mannitol agar, and citrate fermentation using citrate agar.


Background
Enterobacteriaceae are a Family of gram negative non-spore forming rod shaped bacteria. They are found everywhere; in soil, water, and decaying matter and also in guts of animals and humans where they can be pathogens or members of the normal intestinal micro ora 1,2 . Antimicrobial resistance in Enterobacteriaceae is a serious threat to public health due to the association of different resistance mechanisms and the insu cient development of new drugs which make these microorganisms resistant to almost all available antibiotics 3 . Animal husbandry, mainly poultry and pig farming which provide food and jobs to a great number of people in this Region, are highly linked to zoonosis 4,5 . Chicken are an important source of contamination with clinically important human pathogens 6 , they tend to harbour extensive numbers and diverse types of enteric bacteria both commensal and pathogenic 7 . Such co-occurrence coupled with abuse of antibiotics in Cameroonian poultries 8 give opportunity for sharing of resistance genes among species and chromosomal recombinations due to stress from antibiotics leading to emergence of resistant strains, multidrug resistance and emergent zoonoses 7,9,10 . When antimicrobial resistance develops in commensal intestinal micro ora it tends to go unchecked making these commensals to act as reservoir of antimicrobial resistance 7 . Thus the development of antimicrobial resistance in commensal Enterobacteriaceae of animal origin is a measure for early detection of antimicrobial resistance in the community 11 . Within the West Region, there are regular reports of gastrointestinal infections among patients, most of these infections are linked to Enterobacteriaceae 12,13 . Treatment of these diseases has been proving di cult at times due to the development of antibiotic resistance in these bacteria 14 . Chicken being one of the main meat types in Cameroon 15 and an important source of Enterobacteriaceae infection for humans 6 , and bearing in mind that the development of antibiotic resistance in commensal Enterobacteriaceae of animal origin is a measure for early detection of antibiotic resistance in the community 11 , our quest for patterns of antibiotic resistance in Enterobacteriaceae in the West Region of Cameroon prompted us to use chicken as our sample source. This research aimed to bring out information that can attract the attention of stakeholders, including veterinarians, physicians, microbiologists, livestock producers, public health workers and relevant government agencies to the need for basic salvaging measures to curb chicken spread zoonosis. We supposed that due to the inadequacy of published works in this area, there is the development of antibiotic resistance, multidrug resistance (MDR) and extended spectrum beta lactamase (ESBL) production among animal hosts which go unreported.

Evaluation of risks in farms
Samples were collected from 28 farms in 11 locations in 5 out of the 8 Divisions in the West Region.
Evaluating risks, 25 farms out of 28 regularly used antibiotics on their chicken, 138 subjects sampled were more than 30 days old and fell on the category "old" while 137 were 30 days old or less and fell in the category "young", 7 farms out of 28 regularly cleaned feeders and drinkers for the chicken, 12 farms out of 28 gave unsure water to their chicken and 18 out of 28 farms had dirty environment with or without stagnant sewage. Bacterial carriage in terms of different colony types present in culture correlated signi cantly with the 3 environmental risk factors evaluated. Table 1 Information on the locations of farms, number of samples and isolates and the risk factors to which chicken subjects were exposed Using any other source of water apart from pipe-borne water without prior treatment.

Use of antibiotics
Regular use of antibiotics in feed for chicken subjects or have used on the subject.

Sanitation
Conditions of the environment such as litter, stagnant sewage, rearing of animals around poultry, state of workers' restroom.

Feeding hygiene
Cleaning of feeders and drinkers.

Age
≤30 days for category "young" and 30 days for category "old" chicken.  Table 5 Complementary tables for details of the prevalence of antibiotic resistance in various isolates) The association of resistance to the various antibiotic classes is shown in the following diagram.
The diagram shows that the number of ESBL producing isolates that were quinolone resistant was signi cantly lower than the number of isolates that were not quinolone resistant (32/84 against 52/84 respectively) while the number of aminoglycoside isolates that were resistant to quinolones was signi cantly higher than the number of isolates that were not quinolone resistant (43/66 against 23/66).

Association of risk factors among chickens with antibiotic resistance
The outcome "resistance" indicates isolates showing resistance to at least one antibiotic from one class of choice. The development of resistance to at least one antibiotic class correlated signi cantly to the age of the chicken and food hygiene. Isolates had higher risk of developing resistance on exposure to all risks except unsure water. ESBL production correlated signi cantly to age of chicken and isolates had higher risk of developing ESBL on exposure to all risks except use of antibiotics. MDR correlated signi cantly to the age of the chicken and isolates had a higher risk of developing MDR on exposure to all risks except unsure water. The risk "use of antibiotics" was quasi constant thus did not correlate with outcomes. The sanitary conditions in the various poultries visited were average, not up to standard conditions mainly due to the nature of construction and materials used which promoted poor hygiene around the poultries. Some areas did not have pipe borne water thus farmers used well water with doubtful cleanliness. These conditions favoured extensive infection of the animals 16, 17 . The chicken harboured a myriad of Enterobacteriaceae observed by the co-occurrence of an average of 3 different colony types in each sample. This is telling of the possibility of new antibiotic resistance and multidrug resistance to develop by horizontal gene transfer by means of mobile genetic materials such as plasmids or transposons 7,18 . There was a high prevalence of Escherichia (20.56 %) and Salmonella (18.78 %) among the isolates followed by Klebsiella (9.90 %). These are clinically important genera with several species causing various diseases in human 19 . The isolation of members like Shigella even in few numbers is already a cause for concern given the virulence of the bacterium. However, their prevalence is similar to what is found elsewhere in several studies 20,21 . The fairly uniform prevalence of the organisms in locations studied can be explained by the uniform socio-demographic and geographical nature (grass eld highlands) of the area 22 .
Antibiotic susceptibility testing All the isolated Enterobacteriaceae genera tested showed a high level of resistance to amoxicillin (87. There was observed a generally high MDR in all the genera with a general prevalence of 85.53 %. This is similar to the trend seen elsewhere as reported in Leski  This research showed from the analysis of the odds ratios of risks that poor sanitation at poultry site from environment to feeding predisposed chicken rst to high bacterial carriage and secondly predisposed these bacteria to developing resistance, MDR and ESBLs. Though antibiotic use did not correlate as risk to these outcomes, it was because almost all poultries samples used antibiotics regularly making it a statistical constant. This is con rmed by Guetiya et al. 2016 32 . Correlation of these outcomes with "old age" of the chicken could be explained by the long duration of exposure to the risks 33 , the time to get infected or for coinfection to allow horizontal gene transfers. This researched also showed that ESBL producing isolates that were quinolone resistant were signi cantly lower than those that were not quinolone resistant (32/84 against 52/84 respectively) while aminoglycoside isolates that were resistant to quinolones were signi cantly higher than those that were not quinolone resistant (43/66 against 23/66). This suggests that quinolone resistance could not be largely plasmid mediated because plasmid-carried quinolone resistance genes tend to occur with ESBL producing genes creating the opposite scenario 29 . This observation thus suggests a chromosomal DNA based quinolone resistance developed due to antibiotic misuse as noted in this study. Elena López et al. showed that cipro oxacin induced chromosomal recombinations in E coli enabled the bacterium to resist the drug 10 . However given the high association of aminoglycoside resistance to quinolone resistance, it suggests a shared rather than intrinsic mechanism such as plasmid mediated resistance but the lower overall prevalence of resistance to aminoglycosides shows a smaller contribution of this mechanism in the resistance observed 34,35 .

Conclusion
This research showed that sub-standard hygienic conditions at poultries caused heavy Enterobacteriaceae carriage in chicken in the West Region of Cameroon with a fairly uniform distribution of organisms across the area of study. Enterobacteriaceae in this Region show high resistance to Penicillins and Tetracyclines. 3 rd generation Cephalosporins and Aminoglycosides proved to be drug classes of choice against Enterobacteriaceae with low resistance rates. The high resistance observed against quinolones calls for close monitoring of the use of these antibiotics in the community. The high prevalence of MDR and ESBL production especially in clinically important genera like Salmonella, Escherichia and Klebsiella indicate the necessity for stakeholders to put efforts and resources to combat misuse of antibiotics especially in animal farms. Further studies need to be conducted in this area on human subjects to evaluate same risk factors and consequences on the prevalence of resistance, MDR and ESBL production so as to do comparative studies and make more implication of this study on human populations.

Aim
The aim of this study was to contribute towards ameliorating the management and treatment of diseases linked to Enterobacteriaceae in Cameroon; providing epidemiological data on prevailing Enterobacteriaceae and their resistance patterns and risk factors.

Design
The design used was a cross-section design.

Study site
The Sample enrichment and culture The swabs were dissolved in buffered peptone water and incubated for 24hrs at 38°C 38 . The enriched samples were cultured on EMB agar by plate streaking 1 and incubated for 48 hours.

Isolation of isolates and preservation
Isolated colonies on the EMB agar were identi ed based on colony characteristics, picked and conserved in a conservation medium; a mixture of glycerol and Muller Hilton broth at 1 part : 3 parts and stored in a refrigerator 39,40 . Phenotypic characterisation of isolates A total of nine (9) rapid screening biochemical tests 1,18 were performed on the isolates including: glucose and lactose fermentation, and hydrogen sulphide (H 2 S) and gas production using Kligler Iron Agar (KIA), urease activity and indole production using urea broth and Kovacs' reagent, mannitol fermentation and motility using mannitol agar, and citrate fermentation using citrate agar.
Antibiotic susceptibility testing Phenotypic determination of resistance by penicillinase production was tested by combination disc test with amoxicillin/clavulanic acid with comparison to amoxicillin. Extended spectrum beta-lactamase production was tested by observing resistance to amoxicillin followed by resistance to ceftriaxone or cefotaxime 42 , and then the double disc synergy test was conducted by placing the ceftriaxone or cefotaxime 3 cm from amoxicillin/clavulanic acid giving a clear zone of intersection between the two discs 43 .

Interpretation of phenotypic reactions
This research work generated information on Enterobacteriaceae carriage in broilers, identity of Enterobacteriaceae organisms, their prevalence, their antibiotic susceptibility pro les (susceptible, resistant, multidrug resistant and ESBL producing) and, odds ratios and correlations between outcomes (Enterobacteriaceae carriage, prevalence, resistance, MDR and ESBL production) and risks.
Enterobacteriaceae carriage was determined by the count of the different colonies on the Enterobacteriaceaespeci c growth medium (EMB agar) based on colony aspect and morphology. The identity of the organism was gotten by interpreting the colony aspect and the results of the phenotypic reactions using a KIA user guide alongside the identi cation software: Feuille de calcule pour l'identi cation microbienne developed by J-Noël JOFFIN (8 Dec 2007 version). Susceptibility to antibiotics was evaluated using standard values as given by Sigma Aldrich® 44 . MDR was assessed as resistance of an isolate to two or more antibiotics belonging antibiotic classes of choice used against Enterobacteriaceae (cephalosporins, carbapenems, quinolones or aminoglycosides). ESBL production was interpreted from double disc synergy test between amoxicillin/clavulanic acid disc and ceftriaxone or cefotaxime disc.
Simple arithmetic operations and conversions were done using Microsoft Excel sheets while frequencies, prevalence, odds ratios, correlations and diagrams were done using IBM SPSS Statistics 20.

Ethical consideration
This study did not involve active participation of the fowls and sampling by cloacal swabs was a non-invasive procedure.
Working with poultry farms falls under the authorities of the Ministry of Livestock, Fisheries and Animal Industries (MINEPIA), thus an authorisation (authorisation reference number 68/18/L/DREPIA-O/SRAG of 04/06/2018) to sample chicken within the West Region was obtained from the Regional Delegation of MINEPIA which also helped to enable collaboration with its Divisional Delegates and farmers.

Consent for publication
Not applicable