The equivalence trials of Salmonella selective enrichment and isolation procedures in standard methods worldwide

ObjectiveTo evaluate the equivalence of 5 Salmonella selective enrichment procedures in standard methods worldwide.MethodsThe growth rates（PR） and the colony characteristics of 5 selective enrichment broths（SC， TTB， RV， MKTTn， SBG） and 5 selective mediums（CAS， XLD， XLT4， HE， BS） were validated according to the GB4789.28—2013. The sensitivity（SE）， the relative trueness（RT） and the acceptable limit（AL） were estimated on highly contaminated foods through the use of 5 protocols combined with 5 selective enrichment broths and 2 selective mediums according to the method of ISO 16140—2：2016. The sensitivity（Se）， the specificity（Sp）， the positive predictive value（PPV） and the negative predictive value（NPV） were simultaneously determined through the use of 10 compound modes combined with selective enrichment broths and selective mediums according to the classical statistical methods.ResultsThe results showed that SC broth had no efficacy in enriching Salmonella enterica serotype Cholerae and Salmonella enterica serotype Cholerae var. Kunzendorf. The CAS and XLD were significantly better than XLT4， HE and BS in identifying typhoidal Salmonella and nontyphoidal Salmonella as well as the phenotype of hydrogen sulfide.Totally 84 samples（22 pieces of meat， 32 pieces of poultry and 30 pieces of aquatic products） were tested. 25 out of the 84（29.76%） were confirmed positive by 5 protocols， and 16 serotypes from 42 isolates were identified. The SE of 5 protocols were 84.00%， 100.00%， 88.00%，88.00% and 92.00%， respectively， but showed no statistically significant difference between each other. The AL of one protocol （SC+TTB） went beyond the limitation of the ISO 16140—2：2016 due to the ineffectiveness of SC broth. The compound mode of RV-CAS had the best comprehensive benefits， while the Se， the Sp， the PPV and the NPV of which were 68.00%， 95.00%， 85.00% and 88.00%， respectively.ConclusionThe ineffectiveness of SC broth reduces the sensitivity of the detection protocol for Salmonella， which could cause a negative deviation， resulting in the threshold exceeding of the Acceptability Limit （AL）. The combinational selective enrichment broth of RV-CAS is recommended to be the optimal mode， with CAS+XLD being recommended to be the optimal selective medium combination. The process of Salmonella serotyping is suggested to be improved in the standard of GB4789.4 to cope with the multi-serotype Salmonella contamination in food chain and be consistent with the international standardization for food microbiology.