A cross-sectional study was conducted using a random selection of 8 Wards from each district as sampling areas. The study aimed to compare sociological data obtained from questionnaires and laboratory results to find any scientific correlation. The chi-squared test was used to compare questionnaire responses, proportion of isolates, prevalence rates, and antibiotic resistance patterns of S. aureus isolates between the two groups. The sample size for the study was calculated using a confidence level (Zα) of 95%, a prevalence rate of 7.6%, and a maximum error (e) of 5%. The formula used to calculate the sample size was [13]

n = Zα2*P(−)e²

n = 1.962 × 0.076 (1 − 0.05²)

n = 107.909

The questionnaire included questions in both open and closed-ended formats to allow respondents to express themselves freely. This approach ensured that a range of information was obtained, including knowledge on key topics such as the withdrawal period, availability of veterinary services, drug administration, and consumption of milk from an animal under treatment without observing the withdrawal period.

To gather data, questionnaires were completed by willing livestock keepers and milk samples were obtained. From the pooling containers used by households, 25 ml of milk was directly collected. The samples were then transferred into sterile screw-capped falcon tubes and kept cool before being frozen within 8 hours. Each district provided 108 milk samples, with an average of 12 samples per ward.

The milk samples were transferred to the laboratory for identification and isolation of S. aureus, determination of antibiotic drug resistance and antibiotic residues.

The fresh milk samples collected from the households in Kilosa and Morogoro Municipality were immediately cultured on blood agar and MacConkey agar according to the standard methods of the examination. Bacterial culturing was performed by streaking three loopfuls from each milk sample onto labeled media plates. The plates were then incubated at 37˚C for 24 hours before readings were taken and recorded.

Two drops of normal saline were added to a sterilized slide using a sterilized wooden stick. A small sample of test colony was transferred to the slide and a form smear, which was dried by gently heating on a gas burner. A few drops of crystal violet were then added and left for about one minute. The slides were washed in a gentle and indirect stream of tap water for 2 seconds and then flooded with mordant iodine and were left to dry for one minute. Thereafter, the slides were washed again in a gentle and indirect stream of tap water for two seconds, followed by washing the slides by using a decolorizing agent (Acetone-alcohol decolorizer) and left for 15 seconds. Finally, slides were washed in a gentle and indirect stream of tap water until no color appeared and then was dried with absorbent paper ready for observation using microscope.

A small amount of bacterial colony was grown on nutrient agar and incubated at 37˚C for 18 - 24 hours, and then transferred to a surface of a clean, dry glass slide using a loop or sterile wooden stick. This was followed by transferring 3 to 4 drops of 3% hydrogen peroxide on the slide and mixed. A positive reaction was indicated by the formation of bubbles from the culture.

Rabbit plasma was diluted in saline (0.85% NaCl) at a ratio of 1:6; 1 ml of the resulting dilution was transferred into small tubes. Several isolated colonies of test organisms were then transferred into the small tubes to give a milky suspension. The suspension containing tubes were incubated at 35˚C in for 4 hours. Examination for clot formation was done by tilting the tube through 90˚; negative tubes were left at room temperature overnight and re-examined.

Antibiotic resistance test was done by using the agar disk diffusion standard method with Mueller-Hinton agar. The plates were incubated at 37˚C for 24 hours in the incubator. The antibiotics tested included oxacillin (1 mg) (30 mg), cefoxitin (30 mg), amoxicillin (30 mg), Vancomycin (30 mg), ampicillin (10 mg), tetracycline (30 mg) and kanamycin (30 mg). Susceptibility categorization was carried out according to National Committee of Clinical Laboratory Standards (NCCLS) recommendations (Hindler, 2010).

Procedure

Appropriate drug discs were impregnated on the surface of the agar plate using sterile forceps or a needle tip. To ensure complete contact with the agar surface the forceps or needle tip was gently pressed down on each disc. The discs were distributed evenly so that they were no closer than 15 mm from the edge of the plate and no two discs were closer than 24 mm from center to center. The plates were then inverted and placed in a 35˚C - 37˚C incubator for 20 hours. After 20 to 24 hours of incubation each plate was examined, and the diameters of complete inhibition zones were measured using sliding calipers.

Interpretation

Sizes of zones of inhibition were then interpreted by referring to National Committee of Clinical Laboratory Standards (NCCLS) recommendations. A categorization of “s” susceptible implies that an infection due to the strain tested may be expected to respond to the recommended dosage of antibiotic for that type of infection and infecting species. Resistant strains “R” on the other hand are not completely inhibited by concentrations within the therapeutic range. The intermediate “I” is a category including strains which may respond to concentrations attainable by unusually high dosage or in areas. The intermediate category also comprises a “buffer zone,” which should prevent major interpretative discrepancies from small, uncontrolled technical factors.

Detection of antibiotic residues in milk samples was done by using Delvotest kit (FAO 2014). Delvotest is a broad-spectrum microbial inhibition test used specifically to detect the presence of antibiotic residues in dairy products, especially milk, to enable producers to adhere to MRL as required by FAO. A kit for 100 samples is composed of 100 ampoules each containing Bacillus stearothermophilus var. calidolactis, nutrients for growth and bromocreso purple.

Briefly, 0.1 ml of a well-homogenized milk sample was added to each ampoule and covered by adhesive foil before being incubated for at least three hours. Color changes were examined/observed from the underneath of a slantly held ampoule and were interpreted as follows:

Yellow color indicated negative test/results, meaning that the milk does not contain any antibiotics, or the antibiotic concentration is below the detection sensitivity of the test.Purple color indicated positive test/results, meaning that the milk sample contains antibiotics equal and above the detection limit.

Results showed that all livestock keepers use drugs (antibiotics) to treat animals in case of sickness in both study areas (Kilosa and Morogoro Municipality). 27% of the livestock keepers in Kilosa required veterinary personnel to administer drugs, whereas 83% of the farmers in Morogoro Municipality relied on veterinary personnel in administering drugs. Assessment of farmers’ knowledge about withdrawal periods showed that only 14.8% of farmers in Kilosa were aware, compared to 89.8% in Morogoro. Milking practices during and after treatment varied markedly between the two areas. In Kilosa, 44% of respondents reported using the milk for household consumption, 38.9% fed it to calves, 2.8% sold it, 13% discarded it, and 1.9% refrained from milking altogether. In Morogoro, 43.3% of respondents fed the milk to calves, 1.9% sold it, and 63.9% discarded it. Importantly, none of the respondents in Morogoro Municipality reported consuming milk from animals undergoing or immediately after treatment. Availability of veterinary services was low in Kilosa compared to Morogoro Municipality; it was 17% and 77% for Kilosa and Morogoro, respectively, as indicated in Table 1.

Table 1. Questionnaire results from the study area.

About 25% of all milk samples collected from Kilosa were contaminated by Staphylococcus aureus isolates, compared to only 13% of the milk samples collected in Morogoro (Table 2).

Antibiotic resistance profile ofStaphylococcus aureus

Determination of the Antibiotic resistance profile was conducted as shown in Figure 1.

All Staphylococcus aureus isolated from both study areas were entirely resistant to ampicillin, amoxicillin and oxacillin. Furthermore, 82% of Staphylococcus aureus isolates from Kilosa were resistant to cefoxitin, 56% were resistant to vancomycin, and only 26% resistant to Kanamycin. Likewise, in Morogoro Municipality 86% of Staphylococcus aureus isolates were resistant to cefoxitin, followed by vancomycin (57.1%) and 21.4% were resistant to kanamycin (Table 3). Multi-drug resistance (MDR) isolates of Staphylococcus aureus were observed in milk from Kilosa whereby (75.2%) of Staphylococcus aureus were resistant to more than two antibiotics, where samples from Morogoro had (79.1%) which were resistant to more than two drugs as indicated in Table 4.

Figure 1. Inhibition zones of S. aureus around different antibiotic discs.

Table 2.S. aureus isolates from milk samples.

Table 3. Antibiotic resistance patterns of S. aureus isolates.

Key: K: Kanamycin, OX: Oxacillin, FOX: Cefoxitin, AML: Amoxicillin, VA: Vancomycin, AMP: Ampicillin, TET: Tetracycline.

Table 4. Multi-Drug resistance patterns of isolates.

Key: TET: K: Kanamycin, OX: Oxacillin, FOX: Cefoxitin, AML: Amoxicillin, VA: Vancomycin, AMP: Ampicillin, TET: Tetracycline.

About 6% of milk samples from Kilosa contained antibiotic residues, whereas the figure was high up in samples from Morogoro Municipality, as 19% were positive for antibiotic residues, as indicated in Table 5 and Figure 2.

Figure 2. Positive and negative results for antibiotics.

Table 5. Milk samples containing antibiotic residues in each ward.