Title: Effect Of Colony Numbers Selected From Plating Media On Salmonella Serogroup Detection From Naturally Contaminated Chicken Carcasses Authors
Submitted to: International Association for Food Protection
Publication Type: Abstract Only
Publication Acceptance Date: July 12, 2009
Publication Date: July 12, 2009
Citation: Cray, P.J., Cox Jr, N.A., Richardson, L.J., Ladely, S.R., Buhr, R.J. 2009. Effect Of Colony Numbers Selected From Plating Media On Salmonella Serogroup Detection From Naturally Contaminated Chicken Carcasses. International Association for Food Protection. Jul 12-15,2009. Grapevine, TX. T1-12. P-32. Technical Abstract: Introduction: Attribution studies are being undertaken to link Salmonella serotypes associated with human illness to food sources. However, studies have demonstrated that culture techniques often influence the sensitivity and specificity associated with bacterial recovery. Purpose: This study evaluated the influence on serogroups of Salmonella through selection of one to three colonies per plate as affected by culture plating media. Methods: Two replications were performed on commercial broiler carcasses (n=26) directly after defeathering. Carcasses were individually bagged, transported on ice and rinsed with 100ml of buffered peptone for 1 minute. One milliliter of rinsate was placed into both GN Hajna (GN) and Tetrathionate (TET) broths and incubated at 37°C for 24h for GN broth or 48h for TET broth prior to transferring 0.1ml to Rappaport-vassiliadis (RV) media. RV tubes were incubated at 37°C for 24h then streaked onto BG Sulfa (BGS) and XLT-4. Following 24h incubation at 37°C, three presumptive colonies were selected from each of the four plates and numbered in order of selection (first, second or third), confirmed and serogrouped. Results: Overall, 90% (BGS; 47/52) and 94% (XLT-4;49/52) of rinsates were positive. No difference in number of positives was observed when 1, 2, or 3 colonies per plate were picked. By carcass, picking a second colony from each plate produced one additional serogroup 16 times and two additional serogroups twice. Picking a third colony produced one additional serogroup nine more times compared to one or two colony picks and once yielded two additional serogroups. Significance: Picking a single colony may underestimate the diversity of Salmonella serogroups within a sample and demonstrates the need to test multiple colonies to increase accuracy. Further characterization to the serotype level and on samples after chill will add additional information. Prior to use in an attribution study, limitations of culture methods/selection criteria for Salmonella should be established in both human and food laboratories.