Carriage of antibiotic-resistant Escherichia coli among healthy children and home-raised chickens: a household study in a resource-limited setting

Risk factors for extended-spectrum beta-lactamase (ESBL)-producing E. coli carriage among children in a food animal-producing region of Ecuador: A repeated measures observational study

BACKGROUND

The spread of antibiotic-resistant bacteria may be driven by human-animal-environment interactions, especially in regions with limited restrictions on antibiotic use, widespread food animal production, and free-roaming domestic animals. In this study, we aimed to identify risk factors related to commercial food animal production, small-scale or "backyard" food animal production, domestic animal ownership, and practices related to animal handling, waste disposal, and antibiotic use in Ecuadorian communities.

METHODS AND FINDINGS

We conducted a repeated measures study from 2018 to 2021 in 7 semirural parishes of Quito, Ecuador to identify determinants of third-generation cephalosporin-resistant E. coli (3GCR-EC) and extended-spectrum beta-lactamase E. coli (ESBL-EC) in children. We collected 1,699 fecal samples from 600 children and 1,871 domestic animal fecal samples from 376 of the same households at up to 5 time points per household over the 3-year study period. We used multivariable log-binomial regression models to estimate relative risks (RR) of 3GCR-EC and ESBL-EC carriage, adjusting for child sex and age, caregiver education, household wealth, and recent child antibiotic use. Risk factors for 3GCR-EC included living within 5 km of more than 5 commercial food animal operations (RR: 1.26; 95% confidence interval (CI): 1.10, 1.45; p-value: 0.001), household pig ownership (RR: 1.23; 95% CI: 1.02, 1.48; p-value: 0.030) and child pet contact (RR: 1.23; 95% CI: 1.09, 1.39; p-value: 0.001). Risk factors for ESBL-EC were dog ownership (RR: 1.35; 95% CI: 1.00, 1.83; p-value: 0.053), child pet contact (RR: 1.54; 95% CI: 1.10, 2.16; p-value: 0.012), and placing animal feces on household land/crops (RR: 1.63; 95% CI: 1.09, 2.46; p-value: 0.019). The primary limitations of this study are the use of proxy and self-reported exposure measures and the use of a single beta-lactamase drug (ceftazidime with clavulanic acid) in combination disk diffusion tests for ESBL confirmation, potentially underestimating phenotypic ESBL production among cephalosporin-resistant E. coli isolates. To improve ESBL determination, it is recommended to use 2 combination disk diffusion tests (ceftazidime with clavulanic acid and cefotaxime with clavulanic acid) for ESBL confirmatory testing. Future studies should also characterize transmission pathways by assessing antibiotic resistance in commercial food animals and environmental reservoirs.

CONCLUSIONS

In this study, we observed an increase in enteric colonization of antibiotic-resistant bacteria among children with exposures to domestic animals and their waste in the household environment and children living in areas with a higher density of commercial food animal production operations.

Human Colonization with Antibiotic-Resistant Bacteria from Nonoccupational Exposure to Domesticated Animals in Low- and Middle-Income Countries: A Critical Review

Data on community-acquired antibiotic-resistant bacterial infections are particularly sparse in low- and middle-income countries (LMICs). Limited surveillance and oversight of antibiotic use in food-producing animals, inadequate access to safe drinking water, and insufficient sanitation and hygiene infrastructure in LMICs could exacerbate the risk of zoonotic antibiotic resistance transmission. This critical review compiles evidence of zoonotic exchange of antibiotic-resistant bacteria (ARB) or antibiotic resistance genes (ARGs) within households and backyard farms in LMICs, as well as assesses transmission mechanisms, risk factors, and environmental transmission pathways. Overall, substantial evidence exists for exchange of antibiotic resistance between domesticated animals and in-contact humans. Whole bacteria transmission and horizontal gene transfer between humans and animals were demonstrated within and between households and backyard farms. Further, we identified water, soil, and animal food products as environmental transmission pathways for exchange of ARB and ARGs between animals and humans, although directionality of transmission is poorly understood. Herein we propose study designs, methods, and topical considerations for priority incorporation into future One Health research to inform effective interventions and policies to disrupt zoonotic antibiotic resistance exchange in low-income communities.

Transmission routes of antibiotic resistant bacteria: a systematic review

Background

Quantification of acquisition routes of antibiotic resistant bacteria (ARB) is pivotal for understanding transmission dynamics and designing cost-effective interventions. Different methods have been used to quantify the importance of transmission routes, such as relative risks, odds ratios (OR), genomic comparisons and basic reproduction numbers. We systematically reviewed reported estimates on acquisition routes’ contributions of ARB in humans, animals, water and the environment and assessed the methods used to quantify the importance of transmission routes.

Methods

PubMed and EMBASE were searched, resulting in 6054 articles published up until January 1st, 2019. Full text screening was performed on 525 articles and 277 are included.

Results

We extracted 718 estimates with S. aureus (n = 273), E. coli (n = 157) and Enterobacteriaceae (n = 99) being studied most frequently. Most estimates were derived from statistical methods (n = 560), mainly expressed as risks (n = 246) and ORs (n = 239), followed by genetic comparisons (n = 85), modelling (n = 62) and dosage of ARB ingested (n = 17). Transmission routes analysed most frequently were occupational exposure (n = 157), travelling (n = 110) and contacts with carriers (n = 83). Studies were mostly performed in the United States (n = 142), the Netherlands (n = 87) and Germany (n = 60). Comparison of methods was not possible as studies using different methods to estimate the same route were lacking. Due to study heterogeneity not all estimates by the same method could be pooled.

Conclusion

Despite an abundance of published data the relative importance of transmission routes of ARB has not been accurately quantified. Links between exposure and acquisition are often present, but the frequency of exposure is missing, which disables estimation of transmission routes’ importance. To create effective policies reducing ARB, estimates of transmission should be weighed by the frequency of exposure occurrence.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-022-07360-z.

High prevalence of antibiotic resistance in commensal Escherichia coli from healthy human sources in community settings

Antibiotic resistance is a global health crisis that requires urgent action to stop its spread. To counteract the spread of antibiotic resistance, we must improve our understanding of the origin and spread of resistant bacteria in both community and healthcare settings. Unfortunately, little attention is being given to contain the spread of antibiotic resistance in community settings (i.e., locations outside of a hospital inpatient, acute care setting, or a hospital clinic setting), despite some studies have consistently reported a high prevalence of antibiotic resistance in the community settings. This study aimed to investigate the prevalence of antibiotic resistance in commensal Escherichia coli isolates from healthy humans in community settings in LMICs. Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we synthesized studies conducted from 1989 to May 2020. A total of 9363 articles were obtained from the search and prevalence data were extracted from 33 articles and pooled together. This gave a pooled prevalence of antibiotic resistance (top ten antibiotics commonly prescribed in LMICs) in commensal E. coli isolates from human sources in community settings in LMICs of: ampicillin (72% of 13,531 isolates, 95% CI: 65-79), cefotaxime (27% of 6700 isolates, 95% CI: 12-44), chloramphenicol (45% of 7012 isolates, 95% CI: 35-53), ciprofloxacin (17% of 10,618 isolates, 95% CI: 11-25), co-trimoxazole (63% of 10,561 isolates, 95% CI: 52-73), nalidixic acid (30% of 9819 isolates, 95% CI: 21-40), oxytetracycline (78% of 1451 isolates, 95% CI: 65-88), streptomycin (58% of 3831 isolates, 95% CI: 44-72), tetracycline (67% of 11,847 isolates, 95% CI: 59-74), and trimethoprim (67% of 3265 isolates, 95% CI: 59-75). Here, we provided an appraisal of the evidence of the high prevalence of antibiotic resistance by commensal E. coli in community settings in LMICs. Our findings will have important ramifications for public health policy design to contain the spread of antibiotic resistance in community settings. Indeed, commensal E. coli is the main reservoir for spreading antibiotic resistance to other pathogenic enteric bacteria via mobile genetic elements.

A Review of Antimicrobial Resistance in Poultry Farming within Low-Resource Settings

The emergence, spread, and persistence of antimicrobial resistance (AMR) remain a pressing global health issue. Animal husbandry, in particular poultry, makes up a substantial portion of the global antimicrobial use. Despite the growing body of research evaluating the AMR within industrial farming systems, there is a gap in understanding the emergence of bacterial resistance originating from poultry within resource-limited environments. As countries continue to transition from low- to middle income countries (LMICs), there will be an increased demand for quality sources of animal protein. Further promotion of intensive poultry farming could address issues of food security, but it may also increase risks of AMR exposure to poultry, other domestic animals, wildlife, and human populations. Given that intensively raised poultry can function as animal reservoirs for AMR, surveillance is needed to evaluate the impacts on humans, other animals, and the environment. Here, we provide a comprehensive review of poultry production within low-resource settings in order to inform future small-scale poultry farming development. Future research is needed in order to understand the full extent of the epidemiology and ecology of AMR in poultry within low-resource settings.

Prevalence of multidrug-, extensive drug-, and pandrug-resistant commensal Escherichia coli isolated from healthy humans in community settings in low- and middle-income countries: a systematic review and meta-analysis

BACKGROUND

The majority of existing studies aimed at investigating the incidence and prevalence of multidrug-resistance by bacteria have been performed in healthcare settings. Relatively few studies have been conducted in community settings, but these have consistently shown a high prevalence of multidrug-resistant bacteria in low- and middle-income countries (LMICs).

OBJECTIVES

To provide an appraisal of the evidence on the high prevalence of multidrug-, extensive drug-, and pandrug-resistance in commensal Escherichia coli isolates from human sources in community settings in LMICs.

METHODS

Using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines, PubMed, EMBASE, MEDLINE, Web of Science, CINAHL, and Cochrane Library databases were systematically searched with the search string: 'Enterobacteriaceae', OR 'E. coli', OR 'Escherichia coli', AND 'antibiotic resistance', OR 'antimicrobial resistance', OR 'drug-resistance', AND 'prevalence', OR 'incidence', OR 'morbidity', OR 'odds ratio', OR 'risk ratio', OR 'confidence interval', OR 'p-value', OR 'rate'. Data were extracted and proportional meta-analysis was performed using the Freeman-Tukey transformation random effect model.

RESULTS

The prevalence of multidrug-, extensive drug- and pandrug-resistance were extracted from articles that met our inclusion criteria and pooled together after a systematic screening of 9,369 items. The prevalence of multidrug-resistance was 28% of 14,336 total cases of isolates tested, 95% CI: 23-32. Extensive drug-resistance was 24% of 8,686 total cases of isolates tested, 95% CI: 14-36. Lastly, pandrug-resistance was 5% of 5,670 total cases of isolates tested, 95% CI: 3-8.

CONCLUSION

This paper provides an appraisal of the evidence on the high prevalence of multidrug-, extensive drug- and pandrug-resistance by commensal E. coli in community settings in LMICs. Our results call for greater effort to be placed at the community level in the design of new and improved public health policies to counter the global threat of antibiotic-resistant infections and bacteria.

Are Food Animals Responsible for Transfer of Antimicrobial-Resistant Escherichia coli or Their Resistance Determinants to Human Populations? A Systematic Review

The role of farm animals in the emergence and dissemination of both AMR bacteria and their resistance determinants to humans is poorly understood and controversial. Here, we systematically reviewed the current evidence that food animals are responsible for transfer of AMR to humans. We searched PubMed, Web of Science, and EMBASE for literature published between 1940 and 2016. Our results show that eight studies (18%) suggested evidence of transmission of AMR from food animals to humans, 25 studies (56%) suggested transmission between animals and humans with no direction specified and 12 studies (26%) did not support transmission. Quality of evidence was variable among the included studies; one study (2%) used high resolution typing tools, 36 (80%) used intermediate resolution typing tools, six (13%) relied on low resolution typing tools, and two (5%) based conclusions on co-occurrence of resistance. While some studies suggested to provide evidence that transmission of AMR from food animals to humans may occur, robust conclusions on the directionality of transmission cannot be drawn due to limitations in study methodologies. Our findings highlight the need to combine high resolution genomic data analysis with systematically collected epidemiological evidence to reconstruct patterns of AMR transmission between food animals and humans.

Worldwide Phylogenetic Group Patterns of Escherichia coli from Commensal Human and Wastewater Treatment Plant Isolates

Escherichia coli is an important microorganism in the gastrointestinal tract of warm-blooded animals. Commensal populations of E. coli consist of stable genetic isolates, which means that each individual has only one phylogenetic group (phylogroup). We evaluated the frequency of human commensal E. coli phylogroups from 116 people and observed that the majority of isolates belonged to group A. We also evaluated the frequency of phylogroups in wastewater samples and found a strong positive correlation between the phylogroup distribution in wastewater and human hosts. In order to find out if some factors, such as geographical location, and climate could influence the worldwide phylogroup distribution, we performed a meta-analysis of 39 different studies and 24 countries, including different climates, living areas, and feeding habits. Unexpectedly, our results showed no substructuring patterns of phylogroups; indicating there was no correlation between phylogroup distribution and geographic location, climate, living area, feeding habits, or date of collection.

Faecal carriage of antibiotic resistant Escherichia coli in asymptomatic children and associations with primary care antibiotic prescribing: a systematic review and meta-analysis

Background

The faecal reservoir provides optimal conditions for the transmission of resistance genes within and between bacterial species. As key transmitters of infection within communities, children are likely important contributors to endemic community resistance. We sought to determine the prevalence of antibiotic-resistant faecal Escherichia coli from asymptomatic children aged between 0 and 17 years worldwide, and investigate the impact of routinely prescribed primary care antibiotics to that resistance.

Methods

A systematic search of Medline, Embase, Cochrane and Web of Knowledge databases from 1940 to 2015. Pooled resistance prevalence for common primary care antibiotics, stratified by study country OECD status. Random-effects meta-analysis to explore the association between antibiotic exposure and resistance.

Results

Thirty-four studies were included. In OECD countries, the pooled resistance prevalence to tetracycline was 37.7 % (95 % CI: 25.9–49.7 %); ampicillin 37.6 % (24.9–54.3 %); and trimethoprim 28.6 % (2.2–71.0 %). Resistance in non-OECD countries was uniformly higher: tetracycline 80.0 % (59.7–95.3 %); ampicillin 67.2 % (45.8–84.9 %); and trimethoprim 81.3 % (40.4–100 %). We found evidence of an association between primary care prescribed antibiotics and resistance lasting for up to 3 months post-prescribing (pooled OR: 1.65, 1.36–2.0).

Conclusions

Resistance to many primary care prescribed antibiotics is common among faecal E. coli carried by asymptomatic children, with higher resistance rates in non-OECD countries. Despite tetracycline being contra-indicated in children, tetracycline resistance rates were high suggesting children could be important recipients and transmitters of resistant bacteria, or that use of other antibiotics is leading to tetracycline resistance via inter-bacteria resistance transmission.

Electronic supplementary material

The online version of this article (doi:10.1186/s12879-016-1697-6) contains supplementary material, which is available to authorized users.

Unexpected distribution of the fluoroquinolone-resistance gene qnrB in Escherichia coli isolates from different human and poultry origins in Ecuador

Fluoroquinolone resistance can be conferred through chromosomal mutations or by the acquisition of plasmids carrying genes such as the quinolone resistance gene (qnr). In this study, 3,309 strains of commensal Escherichia coli were isolated in Ecuador from: (i) humans and chickens in a rural northern coastal area (n = 2368, 71.5%) and (ii) chickens from an industrial poultry operation (n = 827, 25%). In addition, 114 fluoroquinolone-resistant strains from patients with urinary tract infections who were treated at three urban hospitals in Quito, Ecuador were analyzed. All of the isolates were subjected to antibiotic susceptibility screening. Fluoroquinolone-resistant isolates (FRIs) were then screened for the presence of qnrB genes. A significantly higher phenotypic resistance to fluoroquinolones was determined in E. coli strains from chickens in both the rural area (22%) and the industrial operation (10%) than in strains isolated from humans in the rural communities (3%). However, the rates of qnrB genes in E. coli isolates from healthy humans in the rural communities (11 of 35 isolates, 31%) was higher than in chickens from either the industrial operations (3 of 81 isolates, 6%) or the rural communities (7 of 251 isolates, 2.8%). The occurrence of qnrB genes in human FRIs obtained from urban hospitals was low (1 of 114 isolates, 0.9%). These results suggested that the qnrB gene is more widely distributed in rural settings, where antibiotic usage is low, than in urban hospitals and industrial poultry operations. The role of qnrB in clinical resistance to fluoroquinolones is thus far unknown.

Antimicrobial susceptibility and emerging resistance determinants (blaCTX-M, rmtB, fosA3) in clinical isolates from urinary tract infections in the Bolivian Chaco

BACKGROUND

Bolivia is among the lowest-resourced South American countries, with very few data available on antibiotic resistance in bacterial pathogens. The phenotypic and molecular characterization of bacterial isolates responsible for urinary tract infections (UTIs) in the Bolivian Chaco are reported here.

METHODS

All clinical isolates from UTIs collected in the Hospital Basico Villa Montes between June 2010 and January 2014 were analyzed (N=213). Characterization included susceptibility testing, extended-spectrum beta-lactamase (ESBL) detection, identification of relevant resistance determinants (e.g., CTX-M-type ESBLs, 16S rRNA methyltransferases, glutathione S-transferases), and genotyping of CTX-M producers.

RESULTS

Very high resistance rates were observed. Overall, the lowest susceptibility was observed for trimethoprim-sulphamethoxazole, tetracycline, nalidixic acid, amoxicillin-clavulanic acid, ciprofloxacin, and gentamicin. Of E. coli and K. pneumoniae, 11.6% were ESBL producers. Resistance to nitrofurantoin, amikacin, and fosfomycin remained low, and susceptibility to carbapenems was fully preserved. CTX-M-15 was the dominant CTX-M variant. Four E. coli ST131 (two being H30-Rx) were identified. Of note, isolates harbouring rmtB and fosA3 were detected.

CONCLUSIONS

Bolivia is not an exception to the very high resistance burden affecting many South American countries. Optimization of alternative approaches to monitor local antibiotic resistance trends in resource-limited settings is strongly encouraged to support the implementation of effective empiric treatment guidelines.

Fluoroquinolone resistance and molecular characterization of gyrA and parC quinolone resistance-determining regions in Escherichia coli isolated from poultry

Escherichia coli are a common inhabitant of the gastrointestinal tract of mammals and birds; nevertheless, they may be associated with a variety of severe and invasive infections. Whereas fluoroquinolones (FQ) have been banned in the United States for use in poultry production, the use of these antimicrobials in poultry husbandry is still possible in the European Union, although with some restrictions. The aim of this study was to investigate the FQ resistance of 235 E. coli isolates recovered from chickens and turkeys. Minimum inhibitory concentrations were determined by a microdilution method, whereas mutations in the quinolone resistance-determining regions of the target genes, gyrA and parC, were detected by a PCR-based method. High resistance rates (>60%) were observed for nalidixic acid, flumequine, and difloxacin, whereas resistance to ciprofloxacin, danofloxacin, enrofloxacin, marbofloxacin, and sarafloxacin was less frequently reported (<40%). Sixty-four isolates (27.2%) showed full susceptibility toward the tested FQ, but 57 isolates (24.2%) were resistant to all tested FQ. The remaining 114 E. coli isolates (48.5%) were grouped in 5 different resistance patterns. Isolates resistant only to flumequine or nalidixic acid or both possessed 1 gyrA mutation, whereas isolates with further resistance to enrofloxacin, difloxacin, danofloxacin, and sarafloxacin had in addition 1 or 2 parC substitutions. Two gyrA mutations coupled with 1 substitution in parC were detected in isolates resistant to all tested FQ. The number of mutations and their correlation with the in vitro activity of FQ reflected the currently accepted model, according to which a single gyrA substitution is associated with resistance or decreased susceptibility to older quinolones, whereas further gyrA or parC substitutions are needed for a higher level of resistance.

Antimicrobial resistance among Enterobacteriaceae in South America: history, current dissemination status and associated socioeconomic factors

South America exhibits some of the higher rates of antimicrobial resistance in Enterobactericeae worldwide. This continent includes 12 independent countries with huge socioeconomic differences, where the ample access to antimicrobials, including counterfeit ones, coexists with ineffective health systems and sanitation problems, favoring the emergence and dissemination of resistant strains. This work presents a literature review concerning the evolution and current status of antimicrobial resistance threats found among Enterobacteriaceae in South America. Resistance to β-lactams, fluoroquinolones and aminoglycosides was emphasized along with description of key epidemiological studies that highlight the success of specific resistance determinants in different parts of the continent. In addition, a discussion regarding political and socioeconomic factors possibly related to the dissemination of antimicrobial resistant strains in clinical settings and at the community is presented. Finally, in order to assess the possible sources of resistant bacteria, we compile the current knowledge about the occurrence of antimicrobial resistance in isolates in South American' food, food-producing animals and off-hospitals environments. By addressing that intensive intercontinental commerce and tourism neutralizes the protective effect of geographic barriers, we provide arguments reinforcing that globally integrated efforts are needed to decelerate the emergence and dissemination of antimicrobial resistant strains.

Quinolone resistance in absence of selective pressure: the experience of a very remote community in the Amazon forest

BACKGROUND

Quinolones are potent broad-spectrum bactericidal agents increasingly employed also in resource-limited countries. Resistance to quinolones is an increasing problem, known to be strongly associated with quinolone exposure. We report on the emergence of quinolone resistance in a very remote community in the Amazon forest, where quinolones have never been used and quinolone resistance was absent in 2002.

METHODS

The community exhibited a considerable level of geographical isolation, limited contact with the exterior and minimal antibiotic use (not including quinolones). In December 2009, fecal carriage of antibiotic resistant Escherichia coli was investigated in 120 of the 140 inhabitants, and in 48 animals reared in the community. All fluoroquinolone-resistant isolates were genotyped and characterized for the mechanisms of plasmid- and chromosomal-mediated quinolone resistance.

PRINCIPAL FINDINGS

Despite the characteristics of the community remained substantially unchanged during the period 2002-2009, carriage of quinolone-resistant E. coli was found to be common in 2009 both in humans (45% nalidixic acid, 14% ciprofloxacin) and animals (54% nalidixic acid, 23% ciprofloxacin). Ciprofloxacin-resistant isolates of human and animal origin showed multidrug resistance phenotypes, a high level of genetic heterogeneity, and a combination of GyrA (Ser83Leu and Asp87Asn) and ParC (Ser80Ile) substitutions commonly observed in fluoroquinolone-resistant clinical isolates of E. coli.

CONCLUSIONS

Remoteness and absence of antibiotic selective pressure did not protect the community from the remarkable emergence of quinolone resistance in E. coli. Introduction of the resistant strains from antibiotic-exposed settings is the most likely source, while persistence and dissemination in the absence of quinolone exposure is likely mostly related with poor sanitation. Interventions aimed at reducing the spreading of resistant isolates (by improving sanitation and water/food safety) are urgently needed to preserve the efficacy of quinolones in resource-limited countries, as control strategies based only on antibiotic restriction policies are unlikely to succeed in those settings.