Evaluation of antimicrobial resistance profiles of Salmonella isolates from broiler chickens at slaughter in Alberta, Canada.

Antimicrobial-resistant Salmonella species are threatening to become a serious public health problem. Therefore, surveillance and prudent use of antimicrobials is needed in both the agricultural and human health sectors. The aim of this study was to describe the antimicrobial susceptibility profiles of Salmonella isolates recovered from healthy broiler chickens at slaughter from November 2004 to April 2005. Salmonella isolates recovered from 36 broiler flocks in Alberta, Canada, were serotyped and tested for antimicrobial susceptibility against 15 antimicrobials. Of 272 Salmonella isolates tested, 64.0% were resistant to one or more antimicrobials, 10.0% were resistant to three or more antimicrobials, and 1.8% were resistant to five antimicrobials. All isolates were susceptible to amikacin, amoxicillin-clavulanic acid, ceftiofur, cefoxitin, ceftriaxone, ciprofloxacin, and nalidixic acid. The highest prevalence of resistance was to tetracycline (54.8%), followed by streptomycin (24.2%) and sulfisoxazole (8.4%). The most common multiantimicrobial resistance patterns were to streptomycin-tetracycline (24.3%), streptomycin-sulfisoxazole-tetracycline (6.6%), and ampicillin-streptomycin-sulfisoxazole-tetracycline (3.7%). The strongest associations were observed between resistance to kanamycin and tetracycline (odds ratio = 65.7, P = 0.001) and to ampicillin and sulfisoxazole (odds ratio = 62.9, P = 0.001). Salmonella Hadar and Salmonella Heidelberg were the two most common serovars accounting for 40.4 and 13.6% of the total isolates, respectively. Eighty-one percent and 12.7% of Salmonella Hadar isolates and 62.0 and 8.1% of Salmonella Heidelberg isolates were resistant to 1 or more and three or more antimicrobials, respectively. The flock level prevalence of resistance ranged from 5.6% for trimethoprim-sulfamethoxazole to 83.3% for tetracycline. This study provides baseline information on antimicrobial susceptibility of Salmonella isolates of broiler chickens at slaughter in Alberta that can serve as a benchmark for future research.

Validation of a real-time polymerase chain reaction assay for the detection of Salmonella in crops of broiler chickens.

Salmonella is one of the frequent causes of bacterial foodborne diseases with major public health impact in industrialized countries. Food-producing animals, in particular poultry, are major sources of human salmonellosis. Salmonella is normally found in the gastrointestinal tract of animals and can contaminate the carcass during the slaughtering process. In poultry, crops are also colonized by this pathogen. Crops are more likely to get ruptured during evisceration and contaminate the carcass and therefore present a health risk to consumers. Reducing Salmonella colonization in crops could decrease carcass contamination and is considered a potential preharvest critical control point in poultry production. Furthermore, rapid and reliable diagnostic methods to detect Salmonella are needed to monitor crop colonization to help ensure food safety. However, detection of Salmonella by bacteriological methods is time consuming and labor intensive and is not suitable for routine screening of a large number of samples. Therefore, this study was undertaken to validate a real-time PCR (RPCR) assay for the detection of Salmonella spp. in crop samples of broiler chickens. In total, 997 crop samples (35 spiked, 962 field) were processed by both RPCR and culture. The RPCR correctly identified all spiked crop samples. Out of 962 field crop samples, 100 tested positive by RPCR and 88 tested positive by culture for Salmonella, giving a sample level prevalence of 10.4 % (95% CI: 8.54 to 12.50%) and 9.1% (95% CI: 7.40 to 11.15%), respectively. The agreement beyond chance between RPCR and culture was 92% (P < 0.001) and 100% (P < 0.001) for field and spiked samples, respectively. Compared with culture, the sensitivity and specificity of RPCR were 98.86 and 98.51% for field samples and 100 and 100% for spiked samples, respectively. Where bacterial speciation is required, only the positive samples would be cultured. Therefore, RPCR can be used as a good screening tool for Salmonella spp. in crops by eliminating the time-consuming and labor-intensive culture of negative samples.

Evaluation of associations between feed withdrawal and other management factors with Salmonella contamination of broiler chickens at slaughter in Alberta.

Salmonellosis is one of the most common bacterial foodborne diseases of public health concern in industrialized countries. Poultry products are considered an important source of Salmonella-related foodborne disease in humans. This study was undertaken to evaluate the relationship between various management factors including feed withdrawal and transportation time with Salmonella contamination in crops, ceca, and carcasses of broiler chickens at slaughter in Alberta. Using a two-stage sampling procedure, 30 matched crop and cecal samples before evisceration and an additional 30 neck skin samples after final wash of broiler chickens were collected at slaughter. A questionnaire was administered at the time of sampling to collect information on flock management risk factors. Cecal contents were individually screened with Salmonella-specific real-time PCR to detect positive flocks, and all cecal, crop, and neck skin samples from positive flocks were processed further for Salmonella isolation and characterization. The flock prevalence of Salmonella was 57.1% and within-flock prevalence of Salmonella for positive flocks was 17.2, 8.1, and 53.9% for ceca, crops, and neck skins, respectively. Salmonella Hadar was the most common serovar identified from crops, ceca, and neck skins of broiler chickens tested. Longer transport (P = 0.04 for neck skins) and waiting time in-plant (P = 0.04 for crops, P = 0.03 for ceca) were identified as important risk factors for Salmonella contamination of broiler chickens at slaughter. Salmonella contamination of broiler chickens could potentially be minimized by reducing waiting time in-plant for flocks with longer transport time.