Salmonella Serovars and Antimicrobial Resistance Profiles in Beef Cattle, Slaughterhouse Personnel and Slaughterhouse Environment in Ethiopia
Summary
The present study was undertaken to determine the occurrence, distribution and antimicrobial resistance profiles of Salmonella serovars in slaughter beef cattle, slaughterhouse environment and personnel engaged in flaying and evisceration during slaughtering process. A total of 800 samples (each sample type, n = 100) consisting of swabs from hides, slaughterhouse personnel hands at flaying and evisceration, rumen and caecal contents, mesenteric lymph nodes, carcasses and holding pens were collected. Of the total 100 beef cattle examined, 14% were Salmonella positive in caecal content and/or mesenteric lymph nodes. Of the various samples analysed, Salmonella was detected in 31% of hides, 19% of rumen contents, 8% of mesenteric lymph nodes, 6% of caecal contents, 2% of carcass swabs, 9% of palm swabs taken from the hands of personnel in the slaughterhouse during flaying (7%) and evisceration (2%), and in 12% of holding pen swabs. The Salmonella isolates (n = 87) belonged to eight different serovars of which S. Anatum (n = 54) and S. Newport (19) were the major serovars and both serovars were detected in all sample sources except in carcass swabs. Eighteen of the 87 (20.7%) Salmonella serovars consisting of Newport (n = 14), Anatum (n = 3) and Eastbourne (n = 1) were resistant to one or more antimicrobials. Among the antimicrobial resistant Salmonella serovars, S. Newport was multidrug resistant (15.6%) and exhibited resistance to streptomycin, sulphisoxazole and tetracycline.
Impact
- •
Salmonella enterica serovars are widespread in beef cattle at slaughter, slaughterhouse personnel and environment.
- •
The proportion of samples that were Salmonella positive was higher in hides (31%) and rumen content (19%) compared with other sample sources (caecal content 6%, mesenteric lymph nodes 8%, and carcass swabs 2%).
- •
Eighteen of the 87 (20.7%) Salmonella serovars consisting of S. Newport (n = 14), Anatum (n = 3) and Eastbourne (n = 1) were resistant to one or more antimicrobials tested of which multidrug resistance (MDR) was detected in S. Newport (15.6%) isolates.
Introduction
Salmonella is one of the major foodborne pathogens causing gastroenteritis in humans and infection has been associated with many different food types including beef and beef products (Smerdon et al., 2001). Often, infected animals shed Salmonella in faeces without showing clinical signs. Various stress factors such as those associated with transport of animals from farm to slaughterhouse augments shedding of Salmonella from carrier animals. Food animals such as cattle may carry Salmonella at slaughter and can serve as sources of contamination and provides an opportunity for entry of the pathogen into the food products (McEvoy et al., 2003; Fegan et al., 2005). It was also reported that the most common Salmonella serovars isolated from animal carcasses were also the common serovars found on the corresponding raw ground beef products (Schlosser et al., 2000). This implies that the presence of Salmonella in slaughter cattle and slaughterhouse environment and the potential cross-contamination of carcasses and edible organs can pose food safety hazards (McEvoy et al., 2003).
Further complicating the food safety concerns, an increasing number of the Salmonella serovars have shown resistance to various antimicrobials, which are commonly used both in the veterinary and public health sectors. The high prevalence and dissemination of multidrug-resistant (MDR) Salmonella serovars have become a growing public health concern. Of particular significance is the increasing number of Salmonella isolates that are resistant to clinically important antimicrobial agents such as fluoroquinolones and third-generation cephalosporins, which are used for the treatment of life-threatening disease conditions in humans (Winokur et al., 2001; Gebreyes et al., 2004).
Previous studies conducted in slaughtered cattle and by-products in Ethiopia indicated the presence of various Salmonella enterica serovars, and antimicrobial resistance of Salmonella isolates from apparently healthy cattle was relatively low (Nyeleti et al., 2000; Alemayehu et al., 2003) compared with other food animals (Molla et al., 2004; Aragaw et al., 2007). Salmonella is also considered as one of the major causes of diarrhoea in humans in different parts of the country and an increase in MDR Salmonella (non-typhoid and typhoid) has been reported (Mache et al., 1997; Wolday, 1998; Mache, 2002; Beyene et al., 2009). Information on the status of zoonotic pathogens such as Salmonella in cattle slaughterhouse environment and personnel in commercial abattoirs in Ethiopia is very limited. Therefore, this study was conducted to determine the prevalence and antimicrobial resistance profiles of Salmonella serovars from slaughterhouse personnel, slaughterhouse environment and apparently healthy slaughtered beef cattle in central Ethiopia.
Materials and Methods
Study design
A cross-sectional study was conducted on apparently healthy beef cattle, slaughterhouse personnel and slaughterhouse environment at a commercial slaughterhouse in Debre Zeit, Ethiopia. In the slaughterhouse, 500 to 1500 sheep and goats everyday and about 35 to 50 adult beef cattle are slaughtered every Tuesdays and Saturdays. Unscheduled slaughters are undertaken on other weekdays based on the demand from customers. After arriving at the slaughterhouse, the animals stayed for an average of 24–72 h where feed and water were provided. Animals to be slaughtered the next day were inspected and moved into pens where they spent the night without feed before slaughter. The sample size required for the study was estimated based on the expected prevalence of Salmonella according to Thrusfield (2005). A previous study on Salmonella in a small-scale abattoir in the study area reported a prevalence of 7% (Alemayehu et al., 2003). Therefore using the 7% expected prevalence, 95% confidence interval and 5% type I error, the number of animals required to demonstrate the prevalence of Salmonella representative of the slaughter beef cattle population was estimated to be 100. Animals were selected randomly using the animal’s slaughter order.
Sample collections
Samples were collected from three sources: (1) Slaughter cattle: From each slaughtered animal (n = 100), swab samples were aseptically collected from hides, rumen and caecal contents, mesenteric lymph nodes and carcasses. (2) Slaughterhouse personnel: Swab samples were collected from palms of personnel working in flaying (n = 100) and evisceration (n = 100) sections of the slaughtering plant and (3) Holding pens: Swabs (n = 100) were collected from pens in which the animals spent the night before slaughter.
The samples were collected during 15 different visits from October 2005 to February 2006. Swab samples from hide, rumen and caecal contents, mesenteric lymph nodes and carcass of each selected slaughtered animal and hand swabs of individuals involved in flaying and evisceration during the slaughtering process were collected following slaughter line operations. Selected animals and corresponding carcasses and organs were identified before sample collection. Holding pen swabs were collected from five to seven sites based on the number of holding pens used on that particular day. Approximately 1 m by 1 m area of the pens was swabbed using moistened swabs with buffered peptone water (BPW). Hide swabs were taken aseptically immediately after the animals were stunned and before bleeding. Both sides of the animal and the median line that extended from the inguinal region to the neck were rubbed once from the posterior end to the anterior end of the animal using sterile sponge swabs (80 × 40 mm size) moistened with BPW (AES Laboratories, Cedex, France).
Samples from consenting slaughterhouse personnel (palms of individuals involved in flaying and evisceration) were collected by rubbing the swabs on both hands, inside and outside, immediately after they completed flaying and evisceration procedures respectively. Rumen content samples were collected by cutting through the rumen wall using sterile scalpel blades. Approximately 25 ml of rumen fluid was collected in sterile screw capped universal bottles. About 25 g of cecal contents were collected by puncturing through the caecal wall so that the contents were collected directly into sterile universal bottles. The mesentery, with the lymph nodes attached, was removed from the surrounding structures using sterile scissors and transported to the laboratory in separate sterile plastic bags. Each carcass surface was swabbed just before it entered into the chilling room according to a method described previously (McEvoy et al., 2003). The outer surfaces of each carcass (both sides) were rubbed over once from hindquarter to the forequarters uniformly using sterile sponge swabs. Samples were transported on ice to the laboratory and were processed for Salmonella isolation on the same day. The lymph nodes were aseptically removed from the surrounding tissue, and 25 g was weighed, passed over a flame to disinfect the surface and cut into small pieces on sterile Petri dishes using sterile scalpel blades. Each of the 25 g minced lymph node, caecal and rumen contents were put separately into sterile stomacher bags and 225 ml of BPW was added and homogenized with stomacher (Seward Stomacher 400, London, UK). Whenever samples were <25 g, they were pre-enriched in 1 : 9 W:V ratios in BPW.
Isolation and identification of Salmonella
The isolation and identification of Salmonella were undertaken following conventional cultural methods. Briefly, each processed sample was pre-enriched in BPW (1 : 9) and incubated for 16–20 h at 37°C. From the pre-enrichment broth, 100 μl was transferred into 9.9 ml of Rappaport–Vassilliadis (RV) broth (Difco™; Becton Dickinson, Sparks, MD, USA) and incubated at 42°C for 24 h. A loop full of the inoculum from RV was streaked onto xylose lysine deoxycholate agar (Oxoid, Basingstoke, Hampshire, UK) and brilliant green phenol red lactose sucrose (BPLS) agar (Merck, Darmstadt, Germany) plates and incubated at 37°C for 24 h. Presumptive Salmonella colonies were further characterized using conventional biochemical tests (Quinn et al., 1999; ISO (International Organization for Standardization), 2002). Isolated Salmonella colonies were inoculated onto triple sugar iron agar (TSI) (Difco; Becton Dickinson, Claix, France), lysine iron agar (LIA) (Difco™; Becton Dickinson, France), Simmon’s citrate agar (Difco; Becton Dickinson, France) and urea agar (BBL®; Becton Dickinson, France) slants and incubated at 37°C for 24 h. Biochemically confirmed Salmonella isolates were cultured on brain heart infusion agar (Difco; Becton Dickinson, France) and shipped to the Public Health Agency of Canada, Office International des Épizooties (OIÉ) Reference Laboratory for Salmonellosis, Guelph, Ontario, Canada for serotyping and antimicrobial susceptibility testing.
Serotyping
For serotyping, the somatic (O) antigens of the Salmonella isolates were determined using a slide agglutination test as described by Ewing (1986). The flagellar (H) antigens were identified by using a microtechnique that employs microtitre plates (Shipp and Rowe, 1980). The antigenic formulae of Salmonella serovars as listed by Popoff and Le Minor (2001) were used to name the serovars.
Antimicrobial susceptibility testing
The Salmonella isolates were tested for resistance to a panel of 24 antimicrobials using the agar dilution method as described previously (Poppe et al., 2002; Larkin et al., 2004) at the Public Health Agency of Canada, Salmonella Research Laboratory, Guelph, Ontario, Canada. The following antimicrobials were used: amikacin (Amk); ampicillin (Amp); amoxicillin/cavulanic acid (Amc); apramycin (Apr); carbadox (Car); cephalothin (Cef); ceftiofur (Ctf); ceftriaxone (Cef) ceftriaxone (Cro); cefoxitin (Fox); chloramphenicol (Chl); ciprofloxacin (Cip); florfenicol (Fen); gentamicin (Gen); kanamycin (Kan); nalidixic acid (Nal); neomycin (Neo); nitrofurantoin (Nit); spectinomycin (Spt); streptomycin (Str); sulphisoxazole (Sul); sulphamethoxazole/ trimethoprim (Sxt); tetracycline (Tet); tobramycin (Tob) and trimethoprim (Tmp). The susceptibility/resistance breakpoints for the antimicrobials were based on those specified by the Clinical Laboratory Standards Institute (CLSI), formerly National Committee for Clinical Laboratory Standards (NCCLS) performance standard documents: (NCCLS) M31-A2 (National Committee for Clinical Laboratory Standards (NCCLS), 2002), M31-S1 (National Committee for Clinical Laboratory Standards (NCCLS), 2004) and M100-S15 (National Committee for Clinical Laboratory Standards (NCCLS), 2005). For those antimicrobials in which there are no CLSI interpretative standards, relevant literature was used to determine the breakpoints (Dunlop et al., 1998;Hakanen et al., 1999; Allen and Poppe, 2002a,b; Aaerestrup et al., 2003) and details can be found in our previous reports (Aragaw et al., 2007). ATCC reference strains Escherichia coli ATCC 25922, Pseudomonas aeruginosa ATCC 27853 and Escherichia coli ATCC 35218 (the latter strain for examining susceptibility to amoxicillin/clavulanic acid) were used as quality control strains as per CLSI recommendations.
Data analysis
Data analysis was carried out using the Stata version 7 (Stata Corporation, 2001). Prevalence of Salmonella at a sample and animal level was expressed as percentage with 95% confidence interval. To determine the strength of association, odds ratios and 95% confidence intervals were calculated. The data were analysed by comparing proportions, using Pearson’s chi-square or Fisher’s exact test based on the number of observations.
Results
Of the 100 beef cattle examined, 14% of them had Salmonella in their caecal contents and/or mesenteric lymph nodes. The proportion of samples in which Salmonella was detected was relatively higher in hides (31%; 95% CI: 21.8–40.2) compared with other sample sources (P < 0.05), Table 1. In addition, Salmonella was detected in the following samples: rumen contents (19%: 95% CI: 11.2–26.8), holding pens (12%; 95% 6.3–19.7) and mesenteric lymph nodes (8%; 95% CI: 2.6–13.4) (Table 1). In other sample types included in this study, the total number of samples that were Salmonella positive varied from 2% (carcass and eviscerating personnel hand swabs) to 7% (caecal contents and flaying personnel hand swabs). A significant difference was observed in the overall distribution of Salmonella between hide and flayers’ hands (1-sided Fisher’s exact = 0.03). The Odds ratio (OR) calculated to assess the degree of differences was 6.44. Flayers’ hands which came in contact with Salmonella-positive hides were 6.44 times (OR = 6.44, 95% CI [1.1–35.3]) more likely to be contaminated with Salmonella as compared with those who contacted Salmonella-negative hides.
| Sample type (each n = 100) | Number of samples | 95% confidence intervals | |
|---|---|---|---|
| Positive | Prevalence (%) | ||
| Hide swab | 31 | 31 | 21.8–40.2 |
| Hand swabs (at flaying) | 7 | 7 | 1.91–12.1 |
| Hand swabs (at evisceration) | 2 | 2 | 0.2–7.0 |
| Rumen content | 19 | 19 | 11.2–26.8 |
| Caecal content | 6 | 6 | 1.26–10.74 |
| Mesenteric lymph nodes | 8 | 8 | 2.6–13.4 |
| Carcass | 2 | 2 | 0.2–7.0 |
| Holding pens | 12 | 12 | 6.3–19.7 |
| Total | 87/800 | 10.9 | |
Among the 87 isolates of Salmonella recovered during the study, eight serovars were identified. The type and number of serovars identified from the three sources (slaughter cattle, slaughterhouse personnel and environment) are shown in Table 2. The two most frequently isolated serovars from the three sources were S. Anatum (62.1%) and S. Newport (20.7%). S. Eastbourne and S. Urbana were the only two serovars isolated from carcass swabs, and S. Eastbourne was also detected in hides and mesenteric lymph nodes. A maximum of two different serovars were isolated from same individual animal and six individual animals were detected having two different serovars in different samples: S. Anatum and S. Bredeney (in rumen content and hide); S. Newport and S. Anatum (in caecal and rumen contents, n = 2); S. Newport and S. Eastbourne (in lymph node and rumen contents; in carcass and rumen contents); S. Anatum and S. II: 40:b:- (in caecal and rumen contents). Same serovars were identified from individual animals: S. Anatum (hides and lymph node; hides and rumen contents; rumen and caecal contents, n = 3), S. Newport (hides and rumen contents, n = 5), while in the remaining animals Salmonella serovar was detected in only one sample type from rumen and caecal contents and hide swabs.
| Sample type | No. examined | No. positive | Serovar | No. tested | No. resistant (antimicrobials*) |
|---|---|---|---|---|---|
| Hide | 100 | 31 | S. Anatum | 23 | 3: Str (n = 1), Tet(n = 2) |
| S. Newport | 5 | 2 (Tet, StrSulTet) | |||
| S. Eastbourne | 2 | 1 (Tet) | |||
| S. Bredeney | 1 | – | |||
| Hand swabs | 200** | 9 | S. Anatum | 6 | – |
| S. Newport | 3 | 1 (StrSulTet) | |||
| Rumen content | 100 | 19 | S. Anatum | 12 | – |
| S. Newport | 5 | 5: Tet (n = 4), SulTet (n = 1) | |||
| S. Uganda | 1 | – | |||
| S. II 40:b:- | 1 | – | |||
| Caecal content | 100 | 6 | S. Anatum | 2 | – |
| S. Newport | 2 | 2 (Tet) | |||
| S. Typhimurium | 1 | – | |||
| S. II 40:b:- | 1 | – | |||
| Mesenteric lymph nodes | 100 | 8 | S. Eastbourne | 5 | – |
| S. Anatum | 2 | – | |||
| S. Newport | 1 | 1 (Tet) | |||
| Carcass | 100 | 2 | S. Eastbourne | 1 | – |
| S. Urbana | 1 | – | |||
| Holding pens | 100 | 12 | S. Anatum | 9 | – |
| S. Newport | 3 | 3: Tet (n = 2), StrSulTet(n = 1) | |||
| Total | 800 | 87 (10.8%) | 87 | 18 (26.7%) |
- *For list of abbreviations of antimicrobials, refer to Table 1.
- **From individuals involved in evisceration and flaying, each (n = 100).
Eighteen of the 87 (20.7%) isolates were resistant to one or more of the antimicrobials tested (Table 2) and the remaining 69 (79.3%) were pansusceptible. Of the total eight serovars identified, resistance to one or more antimicrobials was detected in three serovars: S. Newport (14/19), S. Anatum (3/54) and S. Eastbourne (1/8). All other serovars were pansusceptible to all the 24 antimicrobials tested. S. Anatum, the predominant serovar (63.2%), was resistant to two antimicrobials (streptomycin and tetracycline) and S. Newport, the second predominant serovar (20.7%) showed resistance to tetracycline, streptomycin and sulphisoxazole. The resistance of the isolates ranged for one to three antimicrobials and the patterns detected include Tet (14.9%), Str (1.1%), SulTet (1.1%) and (StrSulTet (3.4%), Table 2. The frequency of resistance to tetracycline was higher in S. Anatum, S. Newport, and S. Eastbourne. Isolates belonging to S. Newport showed resistance to streptomycin, sulphisoxazole and tetracycline.
Discussion
In the present study, the proportion of samples in which Salmonella was detected was high in the slaughterhouse environment, slaughterhouse personnel and beef cattle. About 14% of the slaughtered cattle carried Salmonella in caecal contents and/or mesenteric lymph nodes. The detection of Salmonella in caecal contents and mesenteric lymph nodes of slaughtered cattle is of significance in food safety as this can easily result in contamination of carcasses and edible organs. The prevalence of Salmonella in slaughtered cattle was higher than that in previous reports (Nyeleti et al., 2000; Alemayehu et al., 2003). This could be associated with transportation stress, hygienic conditions of holding pens and the time that the animals stayed in the lairage before slaughter.
The proportion of hide samples that were Salmonella positive (31/100) was higher than any of the samples types examined from slaughter cattle. This could be partly attributed to the lairage pens in which 12% (12/100) of the swab samples collected from these pens in which the animals stayed overnight before slaughter was Salmonella positive. The occurrence of Salmonella on the external surfaces of cattle upon entry into a slaughterhouse can serve as an indication of contamination that potentially could be transferred to carcass surfaces during the dehiding process (Bacon et al., 2002). In addition, the presence of Salmonella shedding in batches of animals during transit and passing through the lairage will result in contamination of the external surface of the animals, the slaughterhouse environment and other animals coming in contact with them (Southern et al., 2006). The contamination of hide with Salmonella may also originate from the floor where animals are being stunned and bled and this has been reported in other studies as well (Puyalto et al., 1997; Motsoela et al., 2002). In this study, environmental samples were collected from lairage pens only and we do not have data on the status of Salmonella in the stunning areas of the slaughterhouse. The relatively high prevalence of Salmonella in hides (31%) found in this study suggests the risk of contamination of carcasses, organs and slaughterhouse environment that potentially could be transferred during dehiding and other slaughter processes. We were not able to compare our results with other similar studies in the region as there was no accessible published information. Previous studies undertaken elsewhere on Salmonella in cattle hides indicated different rates, and direct comparisons would be difficult as sampling and analysis methods varied from study to study. For instance, the contamination rate of Salmonella in cattle hides reported in other studies (Puyalto et al., 1997; Fegan et al., 2005) varied from 29 to 68%. In another study, it was reported that the proportion of Salmonella-positive hides in beef cattle was higher in samples taken at the slaughterhouse (84.2%) compared with that of Salmonella in the hides (37.3%) taken from the same animals before shipment (Fluckey et al., 2007). Such increases could be associated with cross-contamination of animals during transport to the slaughterhouse, lairage pen environment or possibly because of other stress factors, which enhance shedding of Salmonella from carrier animals. Even though it was difficult to link contamination of hides with carcass in the present study, contaminated hides are considered as potentially one of the major sources of Salmonella contamination of beef carcasses in cattle slaughterhouses.
The proportion of Salmonella isolates from pre-chill carcasses in our study was 2% (2/100) and other studies indicated different carcass contamination rates: 1.3% (Bacon et al., 2002), 2% (Fegan et al., 2005), 2.8% (Alemayehu et al., 2003), 7.6% (McEvoy et al., 2003) and 9.8% (Nyeleti et al., 2000). These differences could be partly attributed to differences in abattoir facilities, sampling and culturing techniques and the level of hygienic standards maintained by the respective abattoirs. However, the presence of even small numbers of Salmonella in carcass meat and edible organs could lead to contamination of red meat and other beef products. In this study, it was not possible to establish a direct link between carcass contamination and hands of personnel involved in flaying and evisceration as the serovars isolated and antimicrobial resistance profiles detected were different in both sources (carcass versus hands of personnel), Table 2. In addition, the recovery of specific Salmonella serovar from carcasses only and not from any other sources included in this study could also suggest the presence of other possible sources of contamination, which requires further detailed epidemiological studies coupled with genotyping approaches. It should, however, be noted that isolation of Salmonella from hand swabs of individuals in the slaughterhouse personnel involved in flaying and evisceration could be partly associated with the high proportion of Salmonella recovered in hides of slaughter cattle (31%). The calculated odds ratio (OR) indicated that hands of individuals in contact with Salmonella positive hides were 6.44 times (OR = 6.44, 95% CI [1.1–35.3]) more likely to be contaminated with Salmonella in comparison with those Salmonella-negative hide contacts.
Among the Salmonella isolates identified, the most frequently isolated serovar was S. Anatum accounting for 62.1% of the total isolates. This serovar was previously reported in Ethiopia from slaughter cattle (Nyeleti et al., 2000), camels (Molla et al., 2004), goats (Woldemariam et al., 2005) and swine (Aragaw et al., 2007). S. Anatum was the second most commonly isolated serovar in cattle in other studies (Motsoela et al., 2002; Dargatz et al., 2003; Fegan et al., 2005). The second most frequent serovar recovered in our study (20.7%) was S. Newport. Both serovars (Anatum and Newport) were recovered from all sources (cattle, slaughterhouse environment and personnel) except in carcass swabs (Table 2). S. Newport was previously detected in other food animals (Aragaw et al., 2007) and was associated with foodborne outbreak among college students caused by contaminated undercooked eggs in Ethiopia (Assefa et al., 1994). This serovar was previously associated with human foodborne gastroenteritis in the USA and Canada (D’Aoust, 1997; Egorova et al., 2008; Irvine et al., 2009). Report of the Centers for Disease Control and Prevention (CDC) indicated that of 33 348 laboratory-confirmed Salmonella enterica infections with known serovars reported to the CDC in 2005, 6982 (21%) were caused by S. Typhimurium and 3295 (10%) were caused by S. Newport (CDC, 2008). Salmonella Newport outbreaks reported to CDC in the USA were generally associated with the consumption of beef (Zhao et al., 2003).
The majority of the Salmonella isolates (73.3%) irrespective of sample sources were pansusceptible to the tested antimicrobials. Even though information is not available on the amount of veterinary antimicrobials used in Ethiopian cattle and no strict regulations exist on the use of drugs in food animals, cattle farmers use antimicrobials mainly for treatment purposes. Multidrug resistance (resistance to three or more antimicrobials) was detected in S. Newport strains only and exhibited resistance to streptomycin, sulphisoxazole and tetracycline. S. Newport isolated from hides, hand swabs and holding pens showed similar antimicrobial resistance patterns (R type StrSulTet). Another resistance pattern (R type Tet) was exhibited by the same serovar recovered from rumen and caecal contents, mesenteric lymph nodes, hides and holding pens suggesting the possibility of having same source of contamination; however, this requires additional DNA finger printing approaches to determine the clonality of the isolates from the different sources and establish relationship with each other. Among the three predominant serovars detected in this study (S. Anatum, S. Newport and S. Eastbourne), antimicrobial resistance was very low in S. Anatum as well as in S. Eastbourne strains. This was consistent with previous reports, which indicated pansusceptible S. Anatum isolates from various sources including beef cattle (Nyeleti et al., 2000; Molla et al., 2004). S. Newport has been reported as an emerging multidrug resistant Salmonella serovar in both humans (Dunne et al., 2000) and animals (Winokur et al., 2001; Rankin et al., 2002; Dargatz et al., 2003).
In summary, the proportion of samples in which Salmonella was detected was relatively high in slaughter cattle, holding pens and slaughterhouse personnel. The detection of some similar Salmonella serovars with identical resistance patterns among isolates from holding pens, hides, caecal and rumen contents could possibly suggest that the source of contamination in the abattoir probably be associated with carrier animals brought to the abattoir and/or contaminated holding pen in which the animals become infected during the 24–72 h stay before slaughter. However, Salmonella strains showing similar phenotypic characteristics (serotyping and antimicrobial resistance profiles) could also be genotypically different. Thus in addition to phenotyping, genotyping approaches such as the pulsed-field gel electrophoresis and dendogram analysis are required to determine the clonality of strains having similar phenotypic characteristics recovered from the three sources (slaughter cattle, slaughterhouse environment and personnel).
Acknowledgements
The authors thank the management and technical staff of the abattoir for the cooperation and assistance during the study period. B.S. was financially supported by a grant from the Ministry of Agriculture and Rural Development, Ethiopia.
