Keith Melvina
Ross Products Division, Abbott Laboratories Microbiology Research, 625 Cleveland Avenue, Columbus, Ohio 43215-1724, USA.
J Food Prot. 1997 Jun;60(6):682-685. doi: 10.4315/0362-028X-60.6.682.
An automated qualitative enzyme-linked fluorescent immunoassay was compared to a conventional method outlined in the FDA for the detection of salmonellae in artificially contaminated milk, whey, and carbohydrate-based products. The evaluation parameters included sensitivity and specificity using pure cultures of Salmonella typhimurium , Salmonella tennessee , and Citrobacter freundii and mixtures of these species to address the effect of competing microflora. The overall detection rate of the conventional method was 97% compared to a detection rate of 96% for the automated system. The conventional method sensitivity rate was 97% for the detection of pure cultures of Salmonella typhimurium and Salmonella tennessee . The automated system sensitivity rate was 96%. The sensitivity rates in the presence of competing microflora for the conventional method and automated system were 96 and 95% respectively. Both the conventional and automated system specificity rates were 100% when challenged with pure cultures of Citrobacter only. Blackburn et al. (Lett. Appl. Microbiol. 19:32-36, 1994) had previously evaluated the VIDAS (Vitek Immuno Diagnostic Assay System) Salmonella Assay using pure cultures of salmonellae in laboratory media. This study addresses the use of the VIDAS for detecting salmonellae when examining complex food matrices.