Prevalence and antimicrobial resistance phenotypes of Salmonella species recovered at various stages of broiler operations in Hathazari, Bangladesh

Background and Aim: Dissemination of multidrug-resistant (MDR) Salmonella through food chains has serious health implications, with higher rates of morbidity and mortality. Broiler meat remains a major reservoir of Salmonella contamination. The lack of proper hygiene in local broiler operations has, therefore, prompted this research into the assessment of Salmonella contamination in local shops and associated antimicrobial resistance (AMR) phenotypes. Materials and Methods: A total of 55 broiler samples including skin, meat, and swab samples from chopping and dressing sites were included in the study. The samples were collected from broiler shops in Hathazari, Bangladesh, and screened for the presence of Salmonella strains using culture-based methods. The isolates were biochemically characterized and further tested for AMR to eight common antibiotics using the disk diffusion technique. Results: Salmonella contaminations were identified in 29% (16/55) of the broiler samples. Swab samples collected from the chopping sites appeared to be contaminated in higher proportions (33%) than those collected from the dressing areas (25%). On the other hand, the skin samples (50%) were detected with a higher percentage of contamination than the meat samples (25%). All Salmonella isolates showed resistance toward at least one of the eight antibiotics used. Although none of the isolates was resistant to all antibiotics, 18.75% showed resistance to a maximum of seven antibiotics. Resistance to nalidixic acid was most prevalent (87.5%), followed by sulfamethoxazole-trimethoprim (81.25%), erythromycin (81.25%), tetracycline (75%), streptomycin (56.25%), ampicillin-clavulanic acid (50%), chloramphenicol (43.75%), and cefotaxime (18.75%). The resistance patterns of the isolates were found to be highly diverse. The most frequently observed pattern was the following: Ampicillinclavulanic acid-sulfamethoxazole-trimethoprim-nalidixic acid-tetracycline-chloramphenicol-streptomycin-erythromycin. Conclusion: The relatively high prevalence of MDR strains in the samples underlies an urgent need for surveillance and control measures concerning hygiene and antibiotic use in local broiler operations.


Introduction
Salmonella species are opportunistic intracellular pathogens responsible for severe foodborne infections that cause substantial morbidity and mortality worldwide [1]. Salmonella infections account for 93.8 million cases of foodborne illnesses every year, 155,000 of which lead to death [2]. Gastroenteritis is the most common of these illnesses, followed by bacteremia and other severe systemic diseases such as typhoid and typhoid fever [3,4]. The major vehicles for the transmission of Salmonella pathogens to humans are contaminated foods, particularly those of animal origin, such as poultry, swine, cattle, eggs, and dairy products [5,6]. Of these, poultry accounts for a considerably higher percentage of Salmonella outbreaks than any of the other food commodities [7]. Because the consumption of broiler chicken has been on the rise in Bangladesh since the past decade [8], the risk for exposure to Salmonella through the direct contact and consumption of the contaminated meat is a public health concern that requires rigorous surveillance.
The emergence of antimicrobial resistance (AMR) in Salmonella serovars has been another major concern over the past two decades that further worsens the associated health problems. Increasing reports on the identification of multidrug-resistant (MDR) Salmonella isolates from broilers have raised concerns that the treatment of salmonellosis may be compromised. High rates of MDR Salmonella have already been documented in several of the South Asian countries [9,10]. The main factor contributing to the emergence of Salmonella with AMR is the use of antibiotics in broiler feed to promote growth of the chickens and in veterinary medicine to treat Available at www.onehealthjournal.org/Vol.7/No.2/2.pdf bacterial infections in those chickens [2]. The AMR obtained therefrom is acquired through several mechanistic and epidemiological events, including random mutation, plasmid exchange, horizontal gene transfer, and clonal spread of the resistant isolates [11,12]. At present, antibiotic-based treatment for human salmonellosis is the most effective method in clinical practice [13]. However, the continuous exposure and even abuse of antibiotics in broilers have facilitated the emergence and dissemination of antibiotic resistance in Salmonella, which may have serious health implications, considering the potential for treatment failure when cases of salmonellosis require medical intervention, especially in children, older adults, and immunocompromised individuals [14]. The screening of Salmonella for AMR in broiler operations is, therefore, essential for providing information on the magnitude of resistance to plan for the prevention and control of future emergence and spread.
Poultry chicken is the cheapest and most consumed meat product in Bangladesh [15]. However, like several other developing countries, the commercial broiler industry in Bangladesh remains underdeveloped. Chickens raised in broiler farms are sent alive to local retail outlets, wherein several chickens are kept in a relatively small space for days to weeks until sold. After being slaughtered and processed, whole chickens are instantly handed over to the customers in polypacks. All processing in the slaughtering house is generally performed on the same processing line without any cleaning, leading to a possible transfer of bacteria from carcass to carcass. In fact, proper hygiene is rarely maintained in the total process, which poses a high risk of zoonotic disease through the ingestion of food or water contaminated with the animals' feces, direct contact, or consumption of the contaminated meat. Furthermore, the uncontrolled use of antimicrobials, which is relatively a common practice in Bangladesh intended for growth stimulation and disease treatment, further increases the risk for emergence and spread of MDR strains of Salmonella.
Several studies have been conducted on the prevalence of Salmonella in poultry processing plants in a few locations in Bangladesh [16][17][18]. However, a continuing need for increased surveillance on the prevalence and antimicrobial susceptibility of Salmonella strains at local broiler operations in all areas exists. Moreover, for proper understanding and control of contamination, it is important to determine the relative contributions of various contamination points, including handling, chopping, and dressing. Such information is scarce with respect to the local broiler shops in Bangladesh. Hathazari is one of the major towns in Chattogram district, having a moderately high population density (1700 individuals/km 2 ) [19] and several retail broiler shops. However, information regarding the occurrence of Salmonella in these broiler retailers is not available.
Therefore, this study aimed to determine the prevalence of Salmonella at different stages of broiler operations in retail shops in the Hathazari region of Bangladesh and their resistance profiles toward frequently used antibiotics and to provide essential information for reducing the level of contamination with antibiotic-resistant Salmonella.

Ethical approval
Ethical approval was not required for this study. The samples were collected from the retail market.

Study period and location
This study was conducted from March 2018 to January 2019. The samples were collected from retail shops in Hathazari, Bangladesh, and processed for bacteriological examinations in the laboratory of Department of Microbiology, University of Chittagong.

Sample collection
A total of 55 samples, including swab samples from chopping (9) and dressing (4) sites, skin (6) samples, and meat (36) samples, were collected from local broiler shops in Hathazari, Chattogram, Bangladesh. Swab samples were collected in 0.1% buffered peptone water. All samples were transported to the laboratory, maintained at 4°C, and immediately used for further analysis. Meat samples were homogenized using mortar and pestle and dissolved in 90 mL of 0.1% buffered peptone water before enrichment.

Enrichment, isolation, and identification
For the selective enrichment of Salmonella, 100 μL of the processed samples in buffered peptone water was inoculated to Rappaport-Vassiliadis Salmonella Enrichment Medium (HiMedia, Mumbai, India) and incubated at 42°C for 24 h. The enrichment culture was further streaked on Xylose Lysine Deoxycholate selective growth medium (HiMedia) and incubated at 37°C for 24-48 h [20]. Presumptive Salmonella colonies were identified based on morphological and biochemical characterizations, such as colony appearances, Gram staining, indole, catalase, methyl red, Voges-Proskauer, urease, and citrate utilization tests, and growth characteristics on triple sugar iron agar.

Prevalence of Salmonella contamination
To determine the prevalence of Salmonella contamination at various stages of broiler processing and in skin and meat, a total of 55 samples, including 19 swab samples from the skin, chopping bench, and dressing area, and 36 meat samples (Table-1), were collected from local broiler shops in Hathazari, Bangladesh, and screened for the presence of Salmonella using culture-based techniques by growing successively in two selective media, such as the Rappaport-Vassiliadis broth and Xylose Lysine Deoxycholate agar, which are regularly used for Salmonella isolation from food samples. Sixteen of the 55 samples (29%) were found positive for the presence of Salmonella (Table-1 Considering the distribution of Salmonella contamination in processing operations, the samples collected from chopping sites appeared to be contaminated in higher proportions, with Salmonella recovered from 33% of these samples (Table-1), but one-fourth of the samples collected from dressing areas showed Salmonella contamination, suggesting that the dressing sites may carry a comparatively lower risk for pathogenesis. On the other hand, considering the skin and meat samples, a higher risk appeared to be associated with broiler skin, with half of the samples contaminated with Salmonella, whereas a relatively lower percentage of Salmonella contamination was detected in the meat samples, with only 25% found to have positive results.

Antimicrobial susceptibility
The susceptibility and resistance of the Salmonella isolates were tested against eight frequently used antibiotics, and the results are summarized in Table-2. All isolates showed resistance toward at least one of the antibiotics used, whereas all, except the isolate CS 6 , showed multiple resistance (resistance to ≥ three antibiotics). On the other hand, 18.75% of the isolates demonstrated resistance to a maximum of seven of the eight antibiotics, and 25% of the isolates were resistant to six antibiotics. Intermediate susceptibility toward a minimum of one antibiotic was found in 68.75% of isolates. However, none of the Salmonella isolates showed resistance or susceptibility to all eight antibiotics used.
The antimicrobial resistance patterns (ARPs) exhibited by the isolates were found to be highly diverse, with all isolates exhibiting unique ARPs (Table-3

Discussion
The findings of this study indicate a moderately high prevalence (29%) of overall Salmonella contamination at the local broiler retailers in the Hathazari area of Bangladesh. The frequency of Salmonella contamination, in fact, varies over a wide range from one country to another, from place to place in the same country, or even from one farm to another in the same city. Several reports described a prevalence of only 10% or even less in samples associated with broiler operations [23][24][25][26][27][28][29][30]. Some studies even found a prevalence as low as ~1% in broiler samples [31][32][33]. By contrast, the occurrence of a very high rate of Salmonella contamination was likewise found. An overall prevalence of 50% or more was documented in several investigations [34][35][36][37][38]. With respect to studies on Bangladeshi broiler farms, results similar to those reported herein were also described by a few other groups. For example, the prevalence of Salmonella was reported to be between 20% and 40% in samples obtained from commercial broiler farms [16][17][18]. However, other groups also reported contamination that was much lower or higher than this range [39,40]. The variations in Salmonella prevalence between the reports might result from differences in hygiene practice and surveillance programs in each country or differences in experimental design and detection methods employed in the individual investigation.
In the present study, chopping sites appeared to be associated with a higher proportion of contamination than dressing sites. Moreover, skin samples were found with a higher frequency of contamination than processed meat samples. However, studies particularly focusing on dressing or chopping sites are scarce, and more research is needed to confirm if it is a general trend of Salmonella contamination in broiler shops. About 25% of the meat samples examined in this study were detected with Salmonella. Although the contamination frequency is still an issue of concern, a higher prevalence was found in several other countries. For example, 72%, 53.3%, 52.2%, 35.5%-47.7%, 36.0% and 42.3%, and 36.5% contamination of chicken meat were found in Thailand, Vietnam, China, Austria, Korea, and Belgium, respectively [41].
All Salmonella isolates recovered from the broiler samples in this study were resistant to at least one of the 10 antibiotics; except for one isolate, all showed resistance to two or more antibiotics. The average number of antibiotics to which the isolates were resistant is 4.94, which is consistent with the findings of other studies [42,43]. More than 80% of these isolates showed resistance to nalidixic acid (87.5%), sulfamethoxazole-trimethoprim (81.25%), and tetracycline (81.25%). A similar resistance pattern to these antibiotics is often reported in Salmonella species isolated from chicken and other foodstuffs. Sodagari et al. [44], for example, found 92.8%, 81%, and 61.2% prevalence of resistance to nalidixic acid, tetracycline, and sulfamethoxazole-trimethoprim, respectively, in Salmonella associated with retail chicken meat and giblets. In a study on foodstuff and related sources, 70%, 90%, and 80% of Salmonella enterica isolates showed resistance to nalidixic acid, tetracycline, and sulfamethoxazole-trimethoprim, respectively [45]. Indeed, multiple resistance is more or less frequently observed among Salmonella species analyzed in most of the studies performed today [2,46]. The high proportion of Salmonella isolates of animal origin with resistance to multiple antimicrobial agents indicates excessive or uncontrolled use of antimicrobials in animal feed and medicine.
The 16 Salmonella isolates of this study showed 12 different ARPs. Hernandez et al. [47] also found the same number of patterns in S. enterica serovars isolated from chickens in Spain, although the type of pattern was relatively different. A higher or lower number of patterns were, however, observed in other studies. For example, Antunes et al. [48] reported only two resistance patterns in Salmonella from poultry products. The following were the ARPs that were most frequently encountered in this study: Ampicillin-clavulanic acid-sulfamethoxazole-trimethoprim-nalidixic acid-tetracycline-chloramphenicol-streptomycin-erythromycin, demonstrated by three isolates followed by the resistance patterns: Sulfamethoxazole-trimethoprimnalidixic acid-tetracycline-chloramphenicol-streptomycin-erythromycin and ampicillin-clavulanic acid-sulfamethoxazole-trimethoprim-nalidixic acid-tetracycline-streptomycin-erythromycin. The relatively high occurrence of the "sulfamethoxazole-trimethoprim, nalidixic acid, tetracycline, streptomycin, and erythromycin combination" in the resistance patterns may suggest the presence of gene transfer systems, such as bacterial conjugative plasmids or transposable elements, carrying the respective resistant genes [42,49].

Conclusion
The following are the major findings of this study: (i) A moderately high prevalence of Salmonella contamination in the broiler samples, (ii) a high percentage of AMR of the Salmonella isolates, and (iii) a relatively greater frequency of resistance of the isolates to nalidixic acid, sulfamethoxazole-trimethoprim, erythromycin, and tetracycline. These findings suggest the risk of salmonellosis resulting from local broiler operations and from cross-contamination during the purchase of the broiler products, meal preparation, or consumption of undercooked contaminated meat. Therefore, an urgent need to reduce the risk for development and dissemination of Salmonella contamination through regular surveillance and the prudent use of antimicrobial agents during broiler production and processing, in particular, are required.