Counts, Serovars, and Antimicrobial Resistance Phenotypes of Salmonella on Raw Chicken Meat at Retail in Colombia

Abstract

The objective of this study was to determine Salmonella counts, serovars, and antimicrobial-resistant phenotypes on retail raw chicken carcasses in Colombia. A total of 301 chicken carcasses were collected from six departments (one city per department) in Colombia. Samples were analyzed for Salmonella counts using the most-probable-number method as recommended by the U.S. Department of Agriculture, Food Safety Inspection Service protocol. A total of 378 isolates (268 from our previous study) were serotyped and tested for antimicrobial susceptibility. The overall Salmonella count (mean log most probable number per carcass +/- 95% confidence interval) and prevalence were 2.1(2.0 to 2.3) and 37%, respectively. There were significant differences (P < 0.05) by Salmonella levels (i.e., counts and prevalence) by storage temperature (i.e., frozen, chilled, or ambient), retail store type (wet markets, supermarkets, and independent markets), and poultry company (chicken produced by integrated or nonintegrated company). Frozen chicken had the lowest Salmonella levels compared with chicken stored at other temperatures, chickens from wet markets had higher levels than those from other retail store types, and chicken produced by integrated companies had lower levels than nonintegrated companies. Thirty-one Salmonella serovars were identified among 378 isolates, with Salmonella Paratyphi B tartrate-positive (i.e., Salmonella Paratyphi B dT +) the most prevalent (44.7%), followed by Heidelberg (19%), Enteritidis (17.7%), Typhimurium (5.3%), and Anatum (2.1%). Of all the Salmonella isolates, 35.2% were resistant to 1 to 5 antimicrobial agents, 24.6% to 6 to 10, and 33.9% to 11 to 15. Among all the serovars obtained, Salmonella Paratyphi B dT + and Salmonella Heidelberg were the most antimicrobial resistant. Salmonella prevalence was determined to be high, whereas cell numbers were relatively low. These data can be used in developing risk assessment models for preventing the transmission of Salmonella from chicken to humans in Colombia.