Existing evidence on antibiotic resistance exposure and transmission to humans from the environment: a systematic map

Background Antimicrobial resistance (AMR) is predicted to become the leading cause of death by 2050 with antibiotic resistance being an important component. Anthropogenic pollution introduces antibiotic resistant bacteria (ARB) and antibiotic resistance genes (ARGs) to the natural environment. Currently, there is limited empirical evidence demonstrating whether humans are exposed to environmental AMR and whether this exposure can result in measurable human health outcomes. In recent years there has been increasing interest in the role of the environment and disparate evidence on transmission of AMR to humans has been generated but there has been no systematic attempt to summarise this. We aim to create two systematic maps that will collate the evidence for (1) the transmission of antibiotic resistance from the natural environment to humans on a global scale and (2) the state of antibiotic resistance in the environment in the United Kingdom. Methods Search strategies were developed for each map. Searches were undertaken in 13 bibliographic databases. Key websites were searched and experts consulted for grey literature. Search results were managed using EndNote X8. Titles and abstracts were screened, followed by the full texts. Articles were double screened at a minimum of 10% at both stages with consistency checking and discussion when disagreements arose. Data extraction occurred in Excel with bespoke forms designed. Data extracted from each selected study included: bibliographic information; study site location; exposure source; exposure route; human health outcome (Map 1); prevalence/percentage/abundance of ARB/antibiotic resistance elements (Map 2) and study design. EviAtlas was used to visualise outputs. Results For Map 1, 40 articles were included, from 11,016 unique articles identified in searches, which investigated transmission of AMR from the environment to humans. Results from Map 1 showed that consumption/ingestion was the most studied transmission route. Exposure (n = 17), infection (n = 16) and colonisation (n = 11) being studied as an outcome a similar number of times, with mortality studied infrequently (n = 2). In addition, E. coli was the most highly studied bacterium (n = 16). For Map 2, we included 62 studies quantifying ARB or resistance elements in the environment in the UK, from 6874 unique articles were identified in the searches. The most highly researched species was mixed communities (n = 32). The most common methodology employed in this research question was phenotypic testing (n = 37). The most commonly reported outcome was the characterisation of ARBs (n = 40), followed by characterisation of ARGs (n = 35). Other genetic elements, such as screening for intI1 (n = 15) (which encodes a Class 1 integron which is used as a proxy for environmental ARGs) and point mutations (n = 1) were less frequently reported. Both maps showed that research was focused towards aquatic environments. Conclusions Both maps can be used by policy makers to show the global (Map 1) and UK (Map 2) research landscapes and provide an overview of the state of AMR in the environment and human health impacts of interacting with the environment. We have also identified (1) clusters of research which may be used to perform meta-analyses and (2) gaps in the evidence base where future primary research should focus.