Background: Antimicrobial resistance is a worldwide concern, especially in West Africa. Our objective was to study the impact of human presence on the dissemination of extended-spectrum β-lactamase-producing Enterobacterales (ESBL-E) in the environment by sampling feces from various animal species. Methods: ESBL-E fecal colonization was screened in patients from a Beninese hospital during their stay. Secondary sampling of 10-15 animal feces was conducted around the homes of 16 selected colonized patients. Each sample was cultured on selective media and tested for ESBL phenotype. Bacterial identification and antibiotic susceptibility were performed for ESBL-E. Patients' community living spaces were categorized according to the proportion of cumulative housing surface into rural (< 30%), semi-urban (30-70%) and urban (> 70%) areas. Escherichia coli isolates from 10 patient/neighborhood pairs were sequenced. Results: ESBL-E fecal carriage was 92% (56/61) among hospitalized patients and 69.7% (150/215) in animal samples from the community. There was evidence of a gradient of fecal carriage positively related to the proximity to human dwellings (domestic animals 86%, n = 18/21; poultry-pigs 78%, n = 61/78; sheep/goats 66%, n = 63/96; cattle-wild animals 35%, n = 7/20) (p < 0.001). Animals living in urban areas were significantly more frequently colonized than those living in semi-urban or rural areas (94% vs. 65%, p = 0.04 and 94% vs. 54%, p = 0.02, respectively). The predominant ESBL-E genes were blaCTX-M (100% of ESBL-E including 91% of blaCTX-M-15). Forty-six different sequence types (ST) were identified. Human isolates exhibited higher resistance to ciprofloxacin and to cotrimoxazole than animal isolates (45% vs. 17%, p < 0.001, and 86% vs. 60%, p < 0.001, respectively). Twenty-three clonal strains (SNPs < 10) were detected in 9 clusters. No cluster included both human isolates and animal isolates. However, some clonal strains have been isolated in feces separated from each other by more than 40 km. Conclusions: In the area under study, there is significant colonization of both human and animals by ESBL-E, and a major diversity among the isolated strains. Our geographic analysis suggests a human impact on animal colonization, even though the results do not support direct transmission. Concurrently, genomic analysis suggests the possibility of dissemination over significant distances.

Background: Antimicrobial resistance is a worldwide concern, especially in West Africa. Our objective was to study the impact of human presence on the dissemination of extended-spectrum β-lactamase-producing Enterobacterales (ESBL-E) in the environment by sampling feces from various animal species. Methods: ESBL-E fecal colonization was screened in patients from a Beninese hospital during their stay. Secondary sampling of 10-15 animal feces was conducted around the homes of 16 selected colonized patients. Each sample was cultured on selective media and tested for ESBL phenotype. Bacterial identification and antibiotic susceptibility were performed for ESBL-E. Patients' community living spaces were categorized according to the proportion of cumulative housing surface into rural (< 30%), semi-urban (30-70%) and urban (> 70%) areas. Escherichia coli isolates from 10 patient/neighborhood pairs were sequenced. Results: ESBL-E fecal carriage was 92% (56/61) among hospitalized patients and 69.7% (150/215) in animal samples from the community. There was evidence of a gradient of fecal carriage positively related to the proximity to human dwellings (domestic animals 86%, n = 18/21; poultry-pigs 78%, n = 61/78; sheep/goats 66%, n = 63/96; cattle-wild animals 35%, n = 7/20) (p < 0.001). Animals living in urban areas were significantly more frequently colonized than those living in semi-urban or rural areas (94% vs. 65%, p = 0.04 and 94% vs. 54%, p = 0.02, respectively). The predominant ESBL-E genes were blaCTX-M (100% of ESBL-E including 91% of blaCTX-M-15). Forty-six different sequence types (ST) were identified. Human isolates exhibited higher resistance to ciprofloxacin and to cotrimoxazole than animal isolates (45% vs. 17%, p < 0.001, and 86% vs. 60%, p < 0.001, respectively). Twenty-three clonal strains (SNPs < 10) were detected in 9 clusters. No cluster included both human isolates and animal isolates. However, some clonal strains have been isolated in feces separated from each other by more than 40 km. Conclusions: In the area under study, there is significant colonization of both human and animals by ESBL-E, and a major diversity among the isolated strains. Our geographic analysis suggests a human impact on animal colonization, even though the results do not support direct transmission. Concurrently, genomic analysis suggests the possibility of dissemination over significant distances.