The goal of this study was to estimate the accuracy of the invA-gene-based polymerase chain reaction (PCR) and a culture technique based on pre-enrichment with buffered peptone water, three selective enrichment media (selenite, tetrathionate and Rappaport-Vassiliadis broths) and four selective, solid media (Xylose-Lysine-Tergitol-4, Salmonella/Shigella, Hekton-Enteric and MacConkey), for the detection of Salmonella organisms from caecal samples from slaughter pigs. For this purpose a latent-class (Bayesian) approach was used. Two hundred and three slaughtered pigs were used after grouping them into two groups of 96 and 107 animals. Sensitivity (Se) was estimated to be 56% (95% probability interval 40, 76) for culture and 91% (81, 97) for PCR. The specificity (Sp) of the PCR was 88% (80, 95) while the Sp of the culture had been considered 100% in the statistical analysis as all culture-positive samples were confirmed by serotyping. PCR Se was not affected by the Salmonella serotypes present in the samples analysed. Accordingly, a minimum of 25.5% of the pigs was estimated to harbour Salmonella organisms in their faeces. It was concluded that bacteriology on caecal samples alone was a poor diagnostic method, and that the PCR method could be considered a cost-effective alternative to culture in Salmonella monitoring programmes. However, given the moderate Sp of this molecular technique, PCR-positive samples should be further confirmed through bacteriology.

No abstract

The aim of this study was to use the immunomagnetic separation (IMS) test plus a multiplex polymerase chain reaction (m-PCR) assay to detect Salmonella at genus level and also for the identification of Salmonella enterica serovar Typhimurium in bovine diarrhoeic fecal samples. In all, 400 bovine diarrhoeic fecal specimens were examined by conventional bacterial culture, IMS, and m-PCR. For m-PCR assay, four set primers were selected: 139-141, specific for inv-A gene of Salmonella spp and the RfbJ, FliC and FljB, specific for the rfbJ, FliC and fljB genes of Salmonella Typhimurium or other Salmonella serovars with similar antigenic properties. Thirty-three (8.25%) out of the 400 fecal samples were culture positive for Salmonella serovars. Of these, 66.7% (22 of 33) were Salmonella enterica serovar Typhimurium, and 9.1% (three of 33) were serovar Dublin. In the IMS + m-PCR, four amplified product (663, 526, 284 and 183 bp) were found in all specimens that had serovar Typhimurium (4,5,12:i:1,2), they corresponded, respectively, to the rfbJ, fljB, inv-A and Flic genes of this serovar. In serovar Dublin (1,9,12:g,p:-), Georgia (6,7:b:e,n,z(15)) and, Enteritidis (1,9,12;g,m:-) only one PCR product (284 bp) was amplified from the inv-A gene. In serovars Augustenborg (6,7:i:1,2) and Lindenburg (6,8:i:1,2) three positive bands (526, 284 and 183 bp) were amplified corresponding to the fljB, inv-A and Flic genes, respectively. In serovar Virchow (6,7:r:1,2) two amplified products (284 and 526 bp) from the inv-A and FliC genes were observed. In serovar Gloucster (1,4,12(27):i:1,w) three fragments (183, 284 and 663) from the FliC, inv-A and, rfbJ genes respectively, were observed. In the positive control as expected, four PCR products were amplified corresponding to the FliC, inv-A, fljB and rfbJ genes, respectively. In conclusion, the results of this study showed that detection of Salmonella at genus level with universal ST139-141 primers and identification of Salmonella Typhimurium by using specific primers of O4, H(2):1, 2 and H(1) antigens can potentially permit to more readily evaluate fecal and other types of samples for the presence of these organisms. Compared to bacteriological culture the combination of IMS and m-PCR resulted a faster method for the detection and identification of Salmonella at genus and serovar level by using of universal and specific primers.

The goal of this study was to determine the accuracy of a culture technique and the invA-genbased PCR, for the detection of Salmonella spp from caecal samples from slaughtered pigs. For this purpose a Bayesian approach was used. Two hundred and three pigs were used. Animals were grouped into 2 populations: 96 from small farms and 1 07 from large farms. Sensitivity was 56% (95% Credible Interval: 40-76) for culture and 91% (95%CI : 81-97) for PCR. The specificity of the PCR was 88% (95%CI: 80-95). According to these estimates. the percentage of p1gs with Salmonella organisms in the1r faeces at slaughter in this population was at least 25.5%. It IS concluded that bacteriology on caecal samples alone is a poor diagnostic method to carry out studies on the prevalence of salmonellosis in pigs, and that the sensittvity of th1s technique would be probably lower if the procedure is simplified or 1s carried out on non-stressed pig populations. PCR was considered a reliable screening method for the diagnosis of pig salmonellosis but prone to some misclassifications that should be considered if this technique is used.