U.S. flag

An official website of the United States government

Dot gov

Official websites use .gov
A .gov website belongs to an official government organization in the United States.


Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.


Main content area

Diagnostic accuracy of rectoanal mucosal swab of feedlot cattle for detection and enumeration of Salmonella enterica

Getahun E. Agga, Terrance M. Arthur, John W. Schmidt, Rong Wang, Dayna M. Brichta-Harhay
Journal of food protection 2016 v.79 no.4 pp. 531-537
Salmonella enterica, adults, anus, bacterial shedding, cattle, confidence interval, cross-sectional studies, excretion, feces, feedlots, food pathogens, harvesting, human health, immunomagnetic separation, mucosa, risk
Cattle are noted carriers of the foodborne pathogen Salmonella enterica. The perceived need to decrease the potential human health risk posed by excretion of this pathogen has resulted in numerous studies examining the factors that influence cattle shedding of Salmonella. Fecal grab (FG) samples have been the predominant method used to identify cattle colonized or infected with Salmonella, however FG sampling can be impractical in certain situations and rectoanal mucosal swabs (RAMS) are a more convenient sample type to collect. Despite a lack of studies comparing FG and RAMS for the detection and enumeration of Salmonella fecal shedding, RAMS is perceived as less sensitive because a smaller amount of feces is cultured. In a cross-sectional study to address these concerns, paired RAMS and FG samples were collected from 403 adult feedlot cattle approximately 90 days prior to harvest. Samples were processed for Salmonella enumeration (direct plating) and detection (enrichment and immunomagnetic separation). In all, 89.6% of RAMS and 98.8% of FG samples were positive for Salmonella and concordant prevalence outcomes were observed for 90.8% of samples. Mean enumeration values were 3.01 and 3.12 log10 CFU/ml for RAMS and FG respectively. The sensitivity and specificity of RAMS were 91% (95% confidence interval, CI: 87.5-93%) and 100% (95% CI: 48-100%) respectively for Salmonella detection. Furthermore, RAMS Salmonella enumeration was substantially concordant ('c = 0.89; 95% CI: 0.86-0.91) with FG values. We conclude that RAMS are a reliable alternative to FG for assessing cattle Salmonella fecal shedding status, especially for cattle shedding high-levels of Salmonella.