A systematic review of vaccinations to reduce the shedding of Escherichia coli O157 in the faeces of domestic ruminants.

The objective of this study was to conduct a systematic review and meta-analysis of the published literature to evaluate the efficacy of vaccines in reducing faecal shedding of Escherichia coli O157 in ruminants. A systematic search of eight databases and land-grant university research reports using an algorithm adapted from a previous systematic review of pre-harvest interventions against E. coli O157 was conducted to locate all reports of in vivo trials of E. coli O157 vaccines in ruminants published between 1990 and 2010. All located references were screened by two independent reviewers, and data were extracted from all relevant papers, with treatment effect measured in odds ratios. For trials with a faecal prevalence outcome that did not involve mixing of treated and untreated cattle in the same pen, efficacy was explored using random-effects meta-analysis. Funnel plots were used to evaluate publication bias, and random-effects meta-regression was performed to explore heterogeneity. The search located 20 relevant manuscripts which detailed 24 trials and 46 treatment comparisons; all but one trial involved cattle. There were 9 deliberate challenge trials (19 comparisons), and 15 natural exposure trials (27 comparisons). For Type III protein vaccines, there were 9 natural exposure trials detailing 17 comparisons, and meta-analysis of 8 comparisons revealed that vaccine treatment resulted in a statistically significant reduction in E. coli O157 faecal prevalence [odds ratio (OR)=0.38, 95% confidence interval (CI)=0.29, 0.51]. Siderophore receptor and porin protein (SRP) vaccines (three trials/four comparisons) also reduced faecal prevalence (OR=0.42, 95% CI=0.20, 0.61); however, none of the bacterin vaccine trials (n=3, six comparisons) resulted in a statistically significant reduction in prevalence. The results suggest that Type III protein and SRP vaccines significantly reduce faecal shedding in cattle; however, caution should be taken in interpreting the results because of the heterogeneity in the results.