The prevalence of verocytotoxin-producingEscherichia coli(VTEC) andE. coliO157:H7 in beef in Sweden determined by PCR assays and an immuno-magnetic separation (IMS) method

Abstract

The number of cases of enterohaemorrhagicEscherichia coli(EHEC) in Sweden has increased dramatically since 1995, but no source of the infections has been identified. The prevalence of verocytotoxin-producingE. coli(VTEC), includingE. coliO157:H7 in imported and domestic raw beef in Sweden, was determined by PCR assays and immuno-magnetic separation (IMS, Dynal). VTEC was detected by a multiplex PCR directed at sequences on the vt1 and vt2 genes, and the prevalence ofE. coliO157:H7 was estimated by the IMS assay and a SZ-PCR which detectsE. coliO157:H7 (and some strains of O157:NM, O55:H7 and O55:NM). VTEC was detected in 15.055% (57/368) and 1.051% (6/543) of the imported and domestic beef samples, respectively, by the multiplex PCR. E. coliO157:H7 was detected in 0.053% and 2.054% of the imported beef samples by the IMS and SZ-PCR assays, respectively. In domestic beefE. coliO157:H7 was not detected by either method, indicating that the occurrence in Swedish beef was less than 0.056% with a 95% certainty. The overall prevalence of VTEC was estimated to be 4.050%, by adjusting for the relative proportion of imported and domestic beef in Sweden and, similarly, the overall prevalence ofE. coliO157:H7 was 0.0506% (IMS) or 0.055% (SZ-PCR). We propose that a useful strategy to analyse a large number of food samples for VTEC andE. coliO157:H7 is an initial screening of samples using the multiplex VT-PCR in conjunction with enrichment and buoyant density centrifugation, followed by further analyses on the fraction of VT-positive samples using either the IMS method or SZ-PCR.