Prevalence and risk factors for faecal carriage of multidrug resistant Escherichia coli among slaughterhouse workers

The increasing occurrence of antimicrobial-resistant Escherichia coli in human and animal population has become a global public health problem that requires immediate intervention. We aimed to investigate prevalence and risk factors for faecal carriage of drug-resistant E. coli among slaughterhouse workers. We conducted this cross-sectional study among 118 apparently healthy workers in the largest slaughterhouses in Abuja and Lagos from July to December 2020. E. coli was isolated from stool samples of slaughterhouse workers and antimicrobial susceptibility testing performed using the Kirby-Bauer disk diffusion method. Multi-drug resistance (MDR) was defined as resistance to three or more classes of antibiotics. Majority were males: 88.1% (n = 104), aged > 41 years: 28.8% (n = 34), married: 70.3% (n = 83), and were butchers: 53.4% (n = 63). Prevalence of MDR E. coli was 50% (n = 59), highest among butchers compared to slaughterhouse cleaners. Of 75 E. coli isolates identified, 25.3% (n = 19) were ESBL producers; 78.7% (n = 59) were MDR. Keeping animals (p = 0.01); eating at the slaughterhouse (p = 0.03) and collecting waste (p = 0.02) remained independent risk factors for acquiring MDR E. coli. Prevalence of resistant E. coli was highest among butchers and associated with keeping animals at home, eating at work, and waste-collection. Hand-hygiene and responsible use of antibiotics among slaughterhouse workers should be encouraged.

www.nature.com/scientificreports/ Few studies conducted in Nigeria have examined the presence of MDR-E.coli in human patients and foodproducing animals, however, the risk factors for faecal carriage of MDR-E. coli among SHWs is not yet known 16,17 . A recent review has reported that human exposure to environmental sources was an important risk factor for the acquisition of MDR-E. coli 10 . Most slaughterhouse environments in Nigeria are poorly managed with very poor sanitary conditions, hence promoting the spread of resistant pathogens 18 . Nigeria has a growing population of livestock characterized by an extensive system of management making it difficult to track AMU in this food-animal population because of their constant movement in search of food. Besides, there are problems of illegal importation and sale of antimicrobials for use in livestock 19 . It is also practically impossible to estimate the public health risks encountered by SHWs when they share the same environment with livestock and their waste products 2 .
There is increased AMU among livestock in Nigeria with a resultant increase in the development of AMR among enteric pathogens and indicator bacteria such as E. coli. It is therefore imperative to generate data on the resistance profile of indicator bacteria among apparently healthy individuals exposed to livestock by their occupation for proper diagnosis and treatment of bacterial infections.
We hypothesized that livestock harboring MDR-E. coli facilitates the transmission of the pathogen to workers exposed to livestock and the slaughterhouse environment. This study aimed to investigate the prevalence and risk factors for faecal carriage of MDR-E. coli among SHWs to generate baseline data for the establishment of the AMR Animal-Health Surveillance system in Nigeria. Prevalence was determined by antimicrobial susceptibility testing while the risk was accessed based on interviews and multivariate analysis.
Prevalence of MDR E. coli among slaughterhouse workers. Fifty-nine SHWs of the 118 included in the study tested positive for MDR-E. coli giving a prevalence of 50%. A majority of these, 69.5% (n = 41) resided in Abuja and the highest prevalence of 50.9% was among butchers (prevalence OR = 1.00, 95% confidence interval [CI] = 0.37-2.69; p = 0.00) when compared to slaughterhouse cleaners.
All the E. coli isolates were 100% susceptible to cefoxitin, colistin, imipenem, meropenem, and nitrofurantoin. The maximum number of ESBL-E. coli was isolated from butchers: 57.9% (n = 11) when compared to other SHWs. The resistance rates of the E. coli isolates to ciprofloxacin, ampicillin, tetracycline, sulfonamides, trimethoprim, nalidixic acid, cefotaxime, and chloramphenicol were higher among butchers when compared to the rest of the SHWs. All the isolates were, however, sensitive to cefoxitin, colistin, imipenem, meropenem, and nitrofurantoin (Table 2).
In this study, 6.6% (n = 5) of the isolates were susceptible to all the AMDs tested, 4% (n = 3) were resistant to only one AMD while 9.3% (n = 7) were resistant to only two AMDs. Surprisingly one isolate was resistant to 11 different AMDs including those of critical importance for human health. In 72% (n = 54) of the isolates, a MAR index greater than 0.2 was observed (Fig. 2).
Antimicrobial use patterns of slaughterhouse workers. Among 59 SHWs colonized with MDR-E.
coli, 61% (n = 36) reported that they had taken antibiotics in the last month before stool collection. Less than half 39% (n = 23) reported that the last antibiotics they had taken were Ampiclox (a combination of ampicillin and cloxacillin) especially when they had a knife cut. About a third 35.6% (n = 21) of the workers reported that they self-prescribed antimicrobials more than three times in the last three months before sampling. A majority 93.2% (n = 55) reported that they usually purchased antibiotics from a chemist although 89.8% (n = 53) reported that they usually received direction for use from the pharmacist. More than two-thirds 71.2% (n = 42) reported that they usually completed their dosage while 30.5% (n = 18) reported that they stopped their medication whenever they felt better.
Risk factors for MDR-E. coli among SHWs. Occupational hazards reported among the SHWs were quite a few. Nearly half (49.2%; n = 58) of the workers reported that a major hazard they faced was knife cuts while slaughtering the animals. Over half (56.9%; n = 33) reported that they however had to continue processing slaughtered animals even after sustaining a knife cut, hence their wounds get exposed to infection. Nearly half (48.3%; n = 57) reported that they often handle fetuses recovered from slaughtered pregnant animals.
The www.nature.com/scientificreports/ Five factors that were statistically significant at bivariate analysis (p ≤ 0.05) were included in the unconditional logistic regression model used in this study. These factors include age over 41 years, keeping animals at home, eating at the slaughterhouse while working, collecting slaughterhouse waste, and washing hands with soap. There were fewer female workers hence gender was considered as a potential confounder and included in the final model used for analysis. Using a stepwise elimination approach and controlling for gender, all factors remained statistically significant in the final logistic regression model. We identified four independent risk factors and one protective factor. Being ≥ 41 years (adjusted OR [AOR] = 4.47, 95% CI 1.59-12.60); keeping animals at home; (AOR = 4.04, 95% CI 1.48-11.05); eating at the slaughterhouse (AOR = 3.01, 95% CI 1.14-7.91) and collecting slaughterhouse waste (AOR = 10.31, 95% CI 1.54-69.10) remained independent risk factors for faecal carriage of MDR-E. coli among SHWs. Washing hands with soap (AOR = 2.72, 95% CI 1.06-6.96), however, remained a protective factor against MDR-E. coli among SHWs (Table 4).
Unconditional logistic regression analysis showed that SHWs had 4 times the odds of shedding MDR-E. coli in their stool if they were ≥ 41 years of age (p = 0.01), kept animals at home (p = 0.01), and 3 times more likely if they ate at the slaughterhouse while processing slaughtered cattle (p = 0.03). Workers were 10 times more likely to www.nature.com/scientificreports/ shed MDR-E. coli in their stool if they collected slaughterhouse waste (p = 0.02). They were 3 times more likely to be protected from being colonized by MDR-E. coli if they washed their hands with soap and clean water (p = 0.04).

Discussion
To the best of our knowledge, this is the first study to document the prevalence and risk factors for faecal carriage of MDR-E. coli among SHWs in Abuja and Lagos, Nigeria. The present study recorded a high prevalence of 50% for MDR-E. coli among SHWs in both locations when compared with 36.6% and 36.8% observed in the general population respectively 23,24 . A similar study conducted among poultry workers in Abuja reported a lower prevalence of 39.7% for MDR-E. coli 9 . The high prevalence observed in the present study might be due to AMU in animals which have been associated with the development of bacterial resistance in humans exposed to these food-producing animals [25][26][27] . The present study demonstrates that 72% of MDR-E. coli recovered from the SHWs had a MAR index above 0.2. This implies that the isolates may have been recovered from sources with a high frequency of AMU 28 although multiple AMR is usually detected when plasmids encode AMR phenotype 22 . Our findings show a high prevalence of MDR-E. coli but a low prevalence of ESBL-E. coli colonization in the study population possibly because ceftazidime and cefotaxime are not some of the commonly self-prescribed antimicrobials in the present study. This supports the findings from a survey in the Mekong Delta in Vietnam  www.nature.com/scientificreports/   29 . In contrast to our findings, a study conducted in Pakistan reported a much higher ESBL-E. coli prevalence among cattle and healthy individuals 30 . A related study conducted in southern Nigeria also reported a much higher ESBL and MDR-E. coli prevalence among pig SHWs 31 . This study observed that the MDR-E. coli isolates recovered from SHWs were resistant to ciprofloxacin, tetracycline, ampicillin, sulfonamides, trimethoprim, and nalidixic acid. Inappropriate AMU in humans and animals has been documented in the literature to promote a selective increase in some bacterial populations as well as dissemination of resistant strains 32,33 . A higher ciprofloxacin resistance compared to nalidixic acid resistance was observed among MDR-E. coli isolates supporting recent reports of a similar study in southern Nigeria where a high frequency of ciprofloxacin resistance was observed in isolates recovered from pig slaughterhouse workers 31 . A possible explanation for this observation is likely due to qnr genes where qnrS1 is the most common in Africa especially since fluoroquinolone resistance in different regions have different characteristics 34 . This is not surprising as ciprofloxacin is one of the most commonly prescribed quinolones that is affordable and available over the counter in Nigeria 35 . The MDR-E. coli isolates were susceptible to meropenem and imipenem consistent with recent studies conducted among pig slaughterhouse workers in southern Nigeria 31 . This is not surprising as these antimicrobials are normally given as intravenous injections however, resistance started to emerge in poultry and pigs in the EU with VIM and IMP genes detected in E. coli and Salmonella including a few Oxa-48 genes 36 . Acquired MDR does not occur only in pathogenic bacteria but in the microflora of exposed individuals and animals. According to WHO, the important drivers for the colonization of MDR pathogens is overuse and misuse of antimicrobial drugs 5 although studies have reported that occupation, diet, health, sanitation, and cultural traditions are also drivers for AMR 37 . This could probably explain the high prevalence of MDR-E. coli observed in the present study especially since some SHWs reported self-prescription of antibiotics, incomplete dosages, poor hygiene, and sanitation amongst others.
A related study among pig SHWs in Europe reported that the workers were more likely to be colonized by resistant bacteria if they were involved in the early slaughtering processes 38 . This supports our study results as butchers who are involved in these early processes were the most affected possibly because of close contact with gut content at the beginning of slaughtering especially on open slabs.
Factors associated with MDR-E. coli in the present study were age, exposure to animals at home, eating at the slaughterhouse while working, collecting waste, and hand hygiene. While the former are independent risk factors, the latter is a protective factor. A related study done in Nigeria among poultry workers reported that a history of diarrheoa and work exposure of over 10 years were risk factors for feacal carriage of MDR-E. coli 9 however the present study did not observe any association with these factors among SHWs. A similar study done among poultry SHWs confirms that some risks do exist from occupational exposure for the acquisition of MDR-E. coli 39 .
Washing hands with soap was found to be a protective factor against faecal carriage of MDR-E. coli among the workers. According to WHO, lack of clean water and sanitation facilitates the spread of antimicrobial-resistant microbes 5 . It has been documented that washing hands with soap and clean water is very effective in removing faecal bacteria from the hands 40 confirming our study results. The importance of hand hygiene and sanitation should be encouraged in the fight against AMR.
The mechanism of AMR spread from animals to humans is not clear however, literature shows that animals are reservoirs for E. coli found in humans 41 . Our results show that SHWs were colonized by MDR-E. coli resistant to commonly used antibiotics in humans and livestock. In addition, we identified slaughterhouse waste collection as a risk factor for the acquisition of MDR-E. coli among SHWs. Poor sanitary slaughterhouse environment 31,42 and occupational exposure to food animals may have been responsible for this observation. There is a paucity of studies on risk factors for MDR-E. coli in the target population. Some studies reported travel history and diet as risk factors for faecal carriage of MDR-E. coli however the present study did not consider these factors 10,43 .
There are potential limitations in this study that could be addressed in future research. First, the study focused on the phenotypic characterization of the E. coli isolates to determine the resistance profile of the slaughterhouse workers. Secondly, the study determined work exposure factors as well as the antimicrobial use history of the slaughterhouse workers that put them at risk of being colonized with MDR E. coli. Future studies could explore the possible origin of MDR observed among the slaughterhouse workers as well as the genomic characterization of the isolates to identify the genes, determine clonality and perform cluster analysis for traceability. Table 4. Factors found to be significantly associated with MDR E. coli among slaughterhouse workers in Abuja and Lagos, Nigeria 2020 in a logistic regression model. *Independent risk factors for MDR E. coli that remained significant in the logistic regression model. # Only protective factor against MDR E. coli that was significant in the final logistic regression model used for the multivariate analysis. Gender was included in the final model to control for possible confounders. www.nature.com/scientificreports/ Apparently healthy SHWs could harbor and shed MDR E. coli in their stool hence posing as a public health hazard to the general population. Prevalence of resistant E. coli was highest among butchers. Risk factors for faecal carriage of MDR E. coli were age, animal exposure at home, eating at work, and waste collection. Handwashing with soap and clean water was protective against acquiring MDR E. coli among the workers. This study provides additional evidence demonstrating the occupational hazards for people working in slaughterhouses. We recommend that competent authorities organize regular AMR awareness programs targeted at SHWs to mitigate public health risks. Hand hygiene and responsible use of antibiotics among SHWs should be encouraged.

Methods
Study area. This study was conducted in two major cities; Abuja metropolis, North-Central part of Nigeria (9.01145° N, 7.57961° E) with a population of approximately 3 M people and Lagos city, South-Western Nigeria (6.64983° N, 3.32023° E) inhabited by 14 M people, respectively.

Study design.
A cross-sectional study approach was applied among 118 apparently healthy workers who were randomly selected from the two largest slaughterhouses situated in Abuja (84 samples) and Lagos (34 samples) from July to December 2020. All SHWs aged ≥ 18 years, who volunteered were enrolled as study participants. All study participants signed the informed consent form explaining the purpose and benefits of the study before sample collection. An open data kit (ODK) installed on a smartphone was used to administer a structured interviewer-administered questionnaire to SHWs for the data collection on demographic characteristics and exposure factors.
Sampling method. The largest slaughterhouses each in Abuja and Lagos were selected based on the daily volume of livestock slaughter. Using a simple random sampling method, we selected workers from both slaughterhouses for the study. We collected freshly passed stool samples (one sample per worker) using sterile stool containers and transported the samples in cool boxes to the National Reference Laboratory (NRL), Nigeria Centre for Disease Control, Abuja. The samples collected in Abuja were processed within 3 h of sample arrival in the laboratory for the presence of E. coli. Samples collected in Lagos were stored at − 80 °C thereafter they were transported to the NRL, Abuja for processing. Positivity was determined as the presence of MDR-E. coli in a stool sample.

Bacteriological analysis. Isolation of E. coli isolates.
Briefly, we inoculated one gram of stool sample in 9 ml of buffered peptone water (enrichment broth) and incubated at 37 °C for 24 h. Thereafter, a 10 μl loop-full culture from buffered peptone water was streaked onto the selective media, MacConkey agar, and incubated at 37 °C for 24 h. Three well-isolated suspected E. coli colonies were streaked, usually pink to red onto the indicative media, Eosin Methylene Blue agar, and incubated at 37 °C under aerobic conditions for 24 h 20 . Elevated, moist colonies with a greenish metallic sheen, suggestive of E. coli, were sub-cultured onto the non-selective media, Trypticase Soy agar plates, and incubated for 24 h at 37 °C for the isolation of pure cultures. Conventional biochemical tests were performed on all E. coli isolates 20 and confirmed using a commercially available biochemical test strip, Microbact GNB 24E (Oxoid, UK), according to the Manufacturer's instructions.
Multiple antibiotic resistance (MAR) indexes. We calculated the MAR index for each E. coli isolate using the formula MAR = a/b, where 'a' was the number of antibiotics that the isolate was resistant to while 'b' was the total number of antibiotics that the isolates were subjected to 22 .
Risk factor survey. We administered electronic questionnaires to SHWs using ODK collect app on a smartphone to determine the risk factors for MDR E. coli colonization. We assessed the association between demographics, AMU, exposure factors, and acquiring MDR-E. coli. We obtained permission from slaughterhouse www.nature.com/scientificreports/ management before administering the questionnaire to the workers. We pretested the questionnaire at the Dei-Dei slaughterhouse, however; we did not include data obtained from the pretesting in the final analysis.
Statistical analysis. The data collected with ODK collect app was exported as a Microsoft Excel spreadsheet to a computer using the ODK briefcase tool. Data were analyzed using Epi Info version-7 by computing frequencies and proportions. To identify risk factors for MDR-E. coli, we performed bivariate analysis using odds-ratios (OR) as a measure of association. Risk factors identified as statistically significant at bivariate analysis were included in an unconditional logistic regression model used for multivariate analysis at a 5% significance level. All relevant data are within the paper and available as supporting information ( Supplementary Information 1).
Ethics approval. We obtained ethical approval from the Federal Capital Territory (FCT) Health Research Ethics Committee (Approval Number: FHREC/2020/01/40/04-05-20) and sought permission from the management of each slaughterhouse before the study commenced. All study participants signed the written informed consent form detailing the purpose and benefit of the study before sample collection. The study participants were also assured of confidentiality of the information obtained. All procedures were performed in accordance with the ethics committee's guidelines and requirements.

Data availability
All data generated or analyzed during this study are included in this published article [and its supplementary information files]. www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.