​Prevalence and concentration of Escherichia coli serotype O157 and other VTEC in sheep presented for slaughter in Scotland

Summary

SAC Animal Health Group
Low, D

The project successfully sampled sheep at slaughter in Scotland and carried out testing for Escherichia coli O157:H7 and other non-O157 verocytotoxigenic E. coli (VTEC). This research has increased our knowledge of the presence of E. coli O157:H7 in sheep faeces in Scotland and provided information on the presence of non-O157 VTEC strains.

Importantly, the findings are of highly clustered isolations of E. coli O157:H7. Significant differences were identified in E. coli O157:H7 carriage by animal age group; where strains were more commonly isolated from adult and hogg classes of animals than from lambs. Additionally, E. coli O157:H7 was not isolated during the months of January, February or March 2006. The mean E. coli O157:H7 excretion frequency for lambs presented for slaughter in Scotland during April to December was 3% with a 95% confidence interval (0.005%, 12.6%). The mean frequency of E. coli O157:H7 excretion by adults or hoggs presented for slaughter in Scotland during April to December was 10% with a 95% confidence interval (4%, 17%). The crude detection rate of E. coli O157:H7 calculated from the weighted contribution of these means with respect to the relative proportions of adults, hoggs and lambs slaughtered at different times of year allows an estimate of the mean frequency of E. coli O157:H7 excretion. The prevalence of E. coli O157:H7 carriage by slaughter sheep in Scotland is 3.4% with a 95% confidence interval (0.7%, 9.6%). This result is consistent with the crude 2.2% prevalence in sheep faeces reported by Chapman et al. (1997) and the 1.7% described by Paiba et al. (2002). The highest risk for carriage of E. coli O157:H7 by slaughter sheep in Scotland was during the months of July to September and by sheep that were adults or hoggs.

The work found no statistical evidence of any variation in the faecal carriage of E. coli O157:H7 by abattoir (p=0.42) and age and season have a greater impact on prevalence than the regional origin of the flock. Crucially, our findings do not support the hypothesis that human cases of E.coli O157 are higher in any particular Scottish region as a direct consequence of a higher rate of carriage in sheep in that region.

There were 33 animals positive for E. coli O157:H7 and in seven animals the bacterial counts were at 1 x 103 CFU/g or above and with one sample yielding a count of 1.15 x 107 CFU/g of E.coli O157:H7. This demonstration of sheep with bacterial counts in faeces at 1 x 103 CFU/g or above is consistent with results from cattle where animals exist, termed supershedders, which shed high levels of E. coli O157:H7 and are considered an important pre-determinant for the clustering of positives within groups.

In Continental Europe, the most common non-O157 VTEC serogroups causing human disease are E. coli O111, O26, O103 and O145 (WHO, 1998); which have been reported in 11, 11, 7 and 5 countries respectively (Eklund et. al., 2001; Caprioli and Tozzi, 1998). In this study we found very limited carriage in sheep in Scotland of verocytotoxin positive strains amongst the potential VTEC serogroups of O103, O111 or O145.

The estimated mean excretion rate for E. coli O26 by plating to TBX media was 3.4% with a 95% confidence interval of (2%, 5%). The estimated mean excretion frequency of E. coli O26, by plating to rhamnose selective media, was 4% with a 95% confidence interval of (3%, 6%). Thus, in Scotland, sheep appear less likely than cattle to be a source of VTEC O26 strains. It remains unclear why the numbers of E. coli O26 infections in humans in Scotland are so much lower than infections caused by E. coli O157:H7 (Locking et. al., 2006) when it is apparent that there is carriage of E. coli O26 strains in cattle and sheep in Scotland and these strains are potentially virulent for humans.

In conclusion, we have provided a benchmark for shedding of E. coli O157:H7 by sheep in Scotland. Sheep appear less likely than cattle to be a source of verocytotoxigenic E. coli of serogroup O26 strains and we have produced evidence that VTEC strains of serogroups O103, O111 or O145 are of low prevalence or absent from sheep at slaughter. Despite the carriage of E. coli O26 strains in sheep in Scotland that are potentially virulent for humans the results of enhanced surveillance continues to show E. coli O157:H7 strains as the greatest threat to human health. The highest risk for carriage of E. coli O157:H7 by slaughter sheep in Scotland was during the summer months of July to September and by sheep that were adults or hoggs. Crucially, our findings do not support the hypothesis that human cases of E. coli O157:H7 are higher in any particular Scottish region as a direct consequence of a higher rate of faecal carriage in sheep in that region. Evidence was produced that supershedders for E. coli O157:H7 exist in sheep populations, and these animals are presumed to be the reason for the marked clustering of positive samples.

Project Code: S14005