Reasons for difficulties in isolating the causative organism during food-borne outbreak investigations using STEC as a model pathogen: a systematic review, 2000 to 2019

Introduction: Food-borne disease outbreak investigations use epidemiological, microbiological and food chain evidence to identify the implicated food and inform risk management actions. Aims: We used Shiga toxin-producing Escherichia coli (STEC) as a model pathogen to investigate the success of outbreak strain isolation from food or environmental samples during outbreak investigations, and examined the factors influencing the chance of isolation. Methods: We searched for reports of food-borne STEC outbreak investigations worldwide in peer-reviewed and grey literature in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Results: We found a total of 223 outbreaks suitable for inclusion. Food and/or environmental samples were available for testing in 137 investigations, and the outbreak strain was isolated in 94 (42%) of investigations. We found no significant effect of STEC serovar or size of outbreak on likelihood of successful outbreak strain isolation. Isolation success ranged across different implicated commodities from 86% for beef-related outbreaks to 50% for salads and leafy greens. In 20% of outbreaks with samples available for testing, an additional STEC strain was isolated alongside the outbreak strain and in 6.6%, only an alternative STEC strain was isolated. Risk management action was taken on epidemiological evidence alone in 21 incidents. Conclusion: The principal reasons why the outbreak strain was not isolated were lack of sample availability and methodological issues concerned with laboratory isolation. We recommend strategies that could improve the likelihood of isolation including the rapid collection of samples based on epidemiological intelligence.