MATERIALS AND METHODS

Sampling: Three commercial poultry farms were selected from Selangor area of Malaysia where a total of 80 samples from apparently healthy chickens were collected comprising of skin and feather swabs for Staphylococcus aureus, nasal and tracheal swabs for pasteurella and cloacal swabs for E. coli and Salmonella. All samples were collected in January 2011 using sterile swabs.

Isolation and identification of bacterial isolates: Conventional isolation techniques such as growth on selective media, gram staining and biochemical tests were utilised for the identification of the different bacterial isolates. The Staphylococcus aureus isolates from both skin and feather were identified and confirmed based on colony and cell morphology, gram positive staining, positive catalase and coagulase tests and formation of yellow colonies on Mannitol salt agar. E. coli isolates were identified and confirmed using colony and cell morphology, pinkish colonies on MacConkey agar, gram negative staining, indole and methyl red positive, Voges-Proskauer and citrate negative. Salmonella isolates were identified and confirmed using colony and cell morphology, pinkish colonies on brilliant green agar and positive for motility test. The Pasteurella isolated were identified and confirmed based on colony and cell morphology on blood agar, gram negative staining and oxidase positive test.

Antimicrobial sensitivity test: Antimicrobial sensitivity test was monitored with the disc diffusion assay (Kirby-Bauer) recommended by the NCCLS (2000) and CLSI (2010) on Muller Hinton agar (Oxoid, Milan, Italy); with the following 12 antimicrobial agents : Ampicillin 10 μg (AMP 10), Ciprofloxacin 5 μg (CIP 5), Sulphamethoxazole/Trimethoprim 25 μg (SXT 25), Streptomycin 10 μg (S 10), Tetracycline 10 μg (TE 10), Cephalothin 30 μg (KF 30), Erythromycin 15 μg (E 15), Chloramphenicol 30 μg (C 30), Penicillin G 10 units (P 10), Oxacilin 1 μg (OX 1), Clindamycin 2 μg (DA 2) and Neomycin 30 μg (N 30) also obtained from Oxiod (Milan, Italy). Antibiotic agents were selected based on the importance in treatment of the bacterial isolates. The zone of inhibition was interpreted according to National Laboratory Standard Institute (NLSI, 2010). Although multidrug resistance is defined as resistance to ≥3 antibiotics tested (Oteo et al., 2005), resistance to ≥4 antibiotics tested was considered in this study.

RESULTS

A total of 96 Staphylococcus aureus, 48 E. coli, 7 Pasteurella spp. and 6 Salmonella spp. were isolated from the 3 poultry farms (Table 1). Staphylococcus aureus and E. coli were isolated from all the skin and feather and cloacal swabs respectively from all the farms. However, out of the 7 Pasteurella spp. isolated 4 were from tracheal specimens of farm A and 3 from nasal specimens of farm B. Neither the tracheal nor the nasal specimens from farm C yielded positive colonies for Pasteurella. The only 6 Salmonella spp. isolated were from cloacal specimens of farm B, none was isolated from farm A and C (Table 1).

The result of the antibiotic sensitivity tests for the bacterial isolates is presented in Table 2. More than 50% of the Staphylococcus aureus isolates showed resistance to ampicillin, tetracycline, erythromycin, penicillin and clindamycin, where as none of the isolates were resistance to cephalothin. Ninety four percent of E. coli isolates were resistant to tetracycline while on the hand only 33% of the isolates were resistant to cephalothin. All Salmonella isolated were found to be resistant to tetracycline and clindamycin while only 14% were resistant to cephalothin. Although no breakpoint for zone of inhibition was given for Pasteurella, isolates that have no zone of inhibition were considered as resistant. All the bacterial isolates in this study showed very high resistance (83-100%) to tetracycline while all isolates except Salmonella showed low resistance (0-33%) to cephalothin (Table 2).

Table 3 presents the resistance of the bacterial isolates to the number of antibiotics used in the study. Highest numbers of Staphylococcus aureus isolates (17.7%) were resistant to four antibiotic agents while on the other hand only 3.1% of the isolates were resistant to 10 antibiotic agents. A total of 77.2% of Staphylococcus aureus isolates were resistant to 4 or more antibiotic agents. For E. coli highest number of isolates (25%) was resistant to 9 different antibiotic agents while on the other hand lowest number of isolates (4.2%) was resistant to 5 and 12 antibiotic agents.

Table 1:	Source of samples and number of bacteria isolated from selected poultry farms in Selangor, Malaysia
*: Tracheal sample, **: Nasal sample

Table 2:	Antibiotic resistance of bacteria isolated from selected poultry farms in Selangor, Malaysia
NA: Not applicable in the 2010 guidelines

Table 3:	Resistance of bacterial isolates to number of antibacterial agents tested

All E. coli isolates (100%) were resistant to 5 or more antibiotic agents. Pasteurella isolates presented 71.5% resistance to 8 and above antibiotic agents where as 100% of Salmonella isolates were resistant to 6 or more antibiotic agents.

DISCUSSION

The microbial isolates identified in this study include Staphylococcus aureus, E. coli, Pasteurella sp. and Salmonella sp. Staphylococcus aureus and E. coli appeared to be the most prevalent bacterial species isolated. Staphylococcus aureus is known to be easily carried in the nasopharynx, throat, skin, cuts, boils, nails and as such can easily contribute to the normal microflora (Ekhaise et al., 2008). Salmonella is the least bacteria isolated from the chickens and this trend can be attributed to the good Salmonella control programme practiced by most farms as examination of food to detect Salmonella is routinely carried out for food safety and food-borne disease surveillance.

The result of this study revealed the presence multidrug resistant bacteria from chickens. All isolates showed high resistance to tetracycline while on the other hand all except Salmonella spp. showed very low resistance to cephalothin. The result of this study clearly identified cephalothin as a good choice antibiotic for treatment of infection in the study area. Also all E. coli and Salmonella sp. isolated in this study were found to be resistant to 5 or more antibacterial agents tested in this study, a finding which is supported by earlier reports of Overdevest et al. (2011) that drug resistance in Enterobacteriaceae has increased dramatically during the past decade. The increase which was attributed mainly to increased prevalence of extended-spectrum β-lactamase (ESBL)-producing Enterobacteriaceae (Canton et al., 2008; Coque et al., 2008; Hashim et al., 2011) has caused increase in the use of last-resort antimicrobial drugs (i.e., carbapenems).

In addition, these results provide evidence that there is an increased emergence of antibiotic resistance from commensal bacterial isolates, a finding which is in agreement with the earlier reports of Chikwendu et al. (2008) who found increasing emergence of antibiotic resistance phenotypes in both clinically relevant strains and normal commensal microbiota. The presence of multidrug-resistant bacteria significantly limits the treatment options available for these life-threatening infections. This unfortunate trend is happening at a time when the discovery and development of new antibacterial agents has slowed down drastically as reported by Spellberg et al. (2008) and Menghani et al. (2011).

Although the control of antibiotic usage in Malaysia is legislated, the findings of this study indicated a probable indiscriminate use of antibiotics by the poultry farms in Selangor. This is further supported by earlier reports of uncontrolled used of antibiotics in feed and for treatment by some selected pig farms in Malaysia by Ooi et al. (2011). Although the use of antibiotics in human medicine has influenced the emergence of antibiotic-resistant bacteria, the use of antibiotics in animal agriculture has markedly contributed to this problem as indicated in this study. This has raised several concerns related to human and animal health. The types of bacteria detected from the poultry farms investigated in this study are associated with a variety of human infections (Gehanno et al., 2009; Lim et al., 2009; Fadel and Ismail, 2009). Especially the fact that transfer of resistant bacteria between animals and humans through food products has been documented and can pose a threat to public health ((O’Brien, 2002; Angulo et al., 2004; Molbak, 2004). Although antimicrobial resistance rates of bacteria from animals and their products are available for many countries (Bywater et al., 2004; SWARM, 2007) including Malaysia (Getachew et al., 2010), the guidelines issued by WHO recommends continues surveillance and prudent use of antibacterial agents. The findings in this investigation emphasize the importance of studying multiple genera of bacteria from different animals as sources of human exposure to antibiotic resistance strains. Therefore not only that the chickens are at risks, poultry workers and consumers are equally exposed to serious hazards due to multidrug resistance bacteria. This calls for urgent intervention by regulatory agencies to limit the emergence and spread of these bacteria as well as prudent use of antibacterial agents among farmers in Malaysia.