The Animal Care and Use Research Ethics Committee (ACUREC), University of Ibadan, Nigeria, reviewed and approved the protocols of the study (UI-ACUREC/18/0105). Verbal informed consent was obtained from farm owners before sampling chickens.

A cross-sectional study was conducted to determine the occurrence of multidrug-resistant Salmonella in poultry farms in Ibadan, Oyo State, Southwest Nigeria. A multistage sampling technique was used to select the chickens for sampling. First, five local government areas (LGAs) in Ibadan metropolis, namely Ibadan North, Ibadan North East, Ibadan North West, Ibadan South West and Ibadan South East were randomly selected for the study. Next, a total of 10 commercial layer poultry farms (A–J), two from each of the five LGAs, were purposively recruited based on the presence of commercial layer production and farmers’ consent. Afterwards, a single pen from each farm was randomly selected, from which 36 chickens were sampled by systematic random sampling. The population of the poultry farms ranged from 500 to 2000 chickens aged between 15 and 36 weeks at the time of sampling; the breeds raised were Isa Brown and Bovans Nera Black. The farms all operated an intensive management system using battery cages.

Using a prevalence of 3.5% of Salmonella species reported in layers from poultry farms in Ilorin, Kwara State,¹⁸ the sample size was calculated using the formula,¹⁹ n=z2p(1−p)/d2[Eqn 1] where n = minimum sample number, z = confidence interval at 95% (1.96), p = prevalence level of subject (3.5%), and d = level of precision (2%). A sample size of 324 was determined, and a 10% non-contingency was added to make up for non-response, giving a minimum target sample size of 360.

All sanitary and biosecurity protocols were observed by investigators to avoid the spread of pathogens between farms during sample collection. Farm visits were conducted at about 2-week intervals over eight months from 4 April 2018 to 20 November 2018. Each farm was visited once by the investigators and 36 samples were collected during each visit. The investigators visited each farm on dates specified by the farm managers to ensure that no biosecurity measures were compromised. A systematic random sampling was performed to collect samples from chickens within the pens irrespective of age, breed, and health status. A total of 360 cloacal swabs were collected from live chickens (layers) for the detection of Salmonella. Samples from farms A and C were taken from diseased chickens as the farmers refused access to the apparently healthy ones.

Cloacal swab samples were collected by gently rotating the cotton swab within the cloacae, with care taken to avoid contact with other areas to prevent contamination. All samples were transported in sterile 10 mL tubes on ice at 4 °C to the Veterinary Public Health Laboratory of the Department of Veterinary Public Health and Preventive Medicine, University of Ibadan, for bacteriological analysis. No transport medium was used. Samples were processed immediately on arrival at the laboratory. For each sampled chicken, information on the species, breed, age, health status, management system, and recent antibiotics administered was also obtained and documented.

Isolation of Salmonella was carried out according to the International Organization for Standardization standard 6579 guidelines²⁰ for isolation and characterisation of Salmonella species. Briefly, pre-enrichment was conducted by transferring each swab into 10 mL tubes containing buffered peptone water (Oxoid, Basingstoke, United Kingdom). Each tube was then thoroughly homogenised by vortexing and incubated at 37 °C for 18–20 h. Subsequently, 0.1 mL of the pre-enrichment broth culture was inoculated into 10 mL of Rappaport Vassiliadis broth (Oxoid, Basingstoke, United Kingdom) for the selective enrichment of Salmonella and incubated at 42 °C for 24 h. A loopful from the overnight enrichment broth was then inoculated onto xylose lysine deoxycholate agar (Oxoid, Basingstoke, United Kingdom) and brilliant green agar (Oxoid, Basingstoke, United Kingdom), and the plates were incubated at 37 °C for 24 h. To obtain pure colonies, the presumptive Salmonella colonies that appeared red with black centres on xylose lysine deoxycholate agar and pinkish-white or red surrounded by a red halo on brilliant green agar were subcultured on xylose lysine deoxycholate agar plates and incubated at 37 °C for 24 h. The pure cultures were subsequently inoculated on nutrient agar slants, incubated at 37 °C for 24 h and stored in the fridge at 4 °C.

Salmonella isolates were identified based on colony morphology, and further phenotypically confirmed and characterised using Gram staining (negative) and standard laboratory biochemical tests such as indole (negative), methyl red (positive), Voges-Proskeur (negative), citrate (negative), oxidase (negative), catalase (positive), urease (negative), and triple sugar iron agar (acid/alkaline). Also, sugar fermentation tests for glucose (positive), lactose (negative), mannitol (positive), sucrose (negative), and xylose (positive) were performed.

The antimicrobial susceptibility test was carried out using the agar disk diffusion method.²¹ The sensitivity patterns of the cultured isolates were reported for standard antibiotics according to the Clinical Laboratory Standards Institute.²² Briefly, putative Salmonella isolates on nutrient agar slants were subcultured onto nutrient agar plates and incubated for 24 h at 37 °C. Subsequently, a colony from each plate was emulsified into sterile saline to achieve turbidity equivalent to 0.5 McFarland standard (~ 10⁸ CFU/mL). The inoculum was aseptically and uniformly spread on the surface of prepared Muller Hinton agar plates (Oxoid, Basingstoke, United Kingdom). Thereafter, the antibiotics disks (Biomark Laboratories, Pune, Maharashtra, India) were placed aseptically on the inoculated agar surface using sterile forceps and the plates were incubated at 37 °C for 16–18 h. The following antimicrobials were used: tetracycline (10 µg), cotrimoxazole (25 µg), gentamicin (10 µg), cefuroxime (30 µg), chloramphenicol (10 µg), ciprofloxacin (5 µg), meropenem (10 µg) and amikacin (30 µg). The zones of inhibition around each disk were measured and interpreted according to the Clinical Laboratory Standards Institute guidelines. The Escherichia coli American Type Culture Collection 25922 reference strain was used as the control. An isolate was considered multidrug-resistant if resistant to at least one antimicrobial in three or more antimicrobial classes.

Data were analysed using descriptive statistics such as proportions and percentages and presented as a table and figures. A Pearson’s chi-squared (χ²) test and odds ratios were computed to test the association between the presence of Salmonella and age, breed, farm location, and health status of the chickens. The level of significance was set at p ≤ 0.05. All statistical analyses were conducted using the Statistical Package for the Social Sciences version 21 (IBM Corp., Armonk, New York, United States, 2012).

All of the isolates (n = 77; 100.0%) were resistant to cefuroxime. Isolates were also highly resistant to cotrimoxazole (n = 74; 96.1%), chloramphenicol (n = 73; 94.8%), meropenem (n = 72; 93.5%), gentamicin (n = 69; 89.6%) and tetracycline (n = 64; 83.1%), but only moderately resistant to ciprofloxacin (n = 31; 40.3%) (Figure 2).

We were not able to identify the specific serovars of the Salmonella isolates due to non-availability of antisera to perform serotyping. Also, the findings of this study are not generalisable to farms in Ibadan as only the five LGAs in Ibadan metropolis were sampled. Further work, which should involve the detailed sampling of more LGAs and serotyping of Salmonella isolates, is needed to provide more insight into the prevalence of the various serovars circulating in the poultry population in Ibadan, Nigeria.

We report a high prevalence of Salmonella among chickens from poultry farms in Ibadan, Nigeria. Apparently healthy and sick chickens may serve as potential shedders and propagators of drug-resistant Salmonella, with possible implications for consumer health and the environment. Data generated from this study could inform policies or guidelines for the improvement of on-farm hygiene and biosecurity best practices among poultry farmers to reduce the burden of Salmonella in the industry. Also, Nigeria urgently needs well-coordinated or integrated national surveillance systems and networks to monitor antimicrobial use and resistance in animal and human populations and the environment.