Use of pooled samples for the detection of Salmonella in feces by polymerase chain reaction.

Many epidemiological studies of Salmonella rely on conventional bacteriological culture methods to detect Salmonella in fecal samples. These culture-based methods are inefficient for epidemiological studies in populations with a low prevalence of Salmonella. The objective of this study was to optimize a protocol that uses pooled Salmonella enrichment broth cultures of bovine feces and polymerase chain reaction (PCR) for the detection of the invA gene of Salmonella in feces. In one field trial, 196 animals were sampled, and all samples were tested by culture, invA PCR on individual samples, invA PCR on pools of 5 samples, and BAX PCR on individual samples. All assays showed a high agreement on individual samples (kappa > or = 0.75). The invA PCR was run on each of 40 pools and detected 19 of 22 culture-positive pools. In another field trial, 152 samples were taken from 4 dairies, and the invA PCR was performed on pools of 5 samples in addition to bacteriological culture of individual samples. Salmonella was detected in 5 of the 32 pools (7 total positive samples) by both PCR and culture. One pool was PCR-positive but culture-negative. Pooling did not dramatically affect the performance of the invA PCR; most of the culture-positive samples were detected, including all of the samples when there were 4 or more Salmonella colonies on the agar plate. Based on these field trials, invA PCR on pooled samples appears to be an efficient method of Salmonella detection as long as Salmonella loads are not extremely low.