PertanikaJ. Trop. Agric. Sci. 18(1): 1-8(1995) ISSN: 012&6128© Universiti Pertanian Malaysia Press

Antimicrobial Resistance of E. coli Isolates from Pig Farm Workers,Nondiarrhoeic and Diarrhoeic Pielets

P.Y. CHOO, A. SALAM ABDULLAH and A.R. CHE NYONYA1

Department of Animal Science,Faculty of Veterinary Medicine and Animal Science

Universiti Pertanian Malaysia,43400 Serdang, Selangor, Malaysia

!Department of Biochemistry and MicrobiologyFaculty of Science and Environmental Studies

Universiti Pertanian Malaysia,43400, Serdang, Selangor, Malaysia

Keywords: E. coli, antimicrobial resistance, pig, human, diarrhoea

ABSTRAK

Kajian kerintangan antibiotik ke atas 274 pencilan E. coli daripada pekerja ladang, anak babi yang tidak ciritbirit dan anak babi yang mengalami cirit-birit terhadap 11 agen antimikrob telah dijalankan untuk menentukanbentuk resistan antibiotik E. coli. Lima ladang babi terlibat dalam kajian int. Kerintangan antimikrob adalahlebih tinggi serta ketara (P<0.05) dan hampir 100% bagi sulphasoxazole, streptomycin dan tetracycline padapencilan babi bila dibandingkan dengan pencilan manusia. Purata peratusan kerintangan antibiotik di antaraladang tidak menunjukan perbezaan yang ketara (P>0.05) tidak kira sumber pencilannya, Analisis statistikmenunjukan bahawa kerintangan antibiotik adalah rendah serta ketara (P<0.05) bagi pencilan dari pekerjaladang bila dibandingkan dengan pencilan-pencilan dari anak babi yang tidak cirit birit dan yang mengalamicirit-birit Peratusan pencilan yang kerintangan terhadap sekurang-kurannya 8 antibiotik ialah 27.7, 5.6 dan1.0 masing-masing bagi babi yang cirit-birit, babi yang tidak cirit birit dan pekerja ladang, Kajian ini jugamenunjukan bahawa purata kerintangan antimikrob adalah lebih tinggi bagi ladang yang besar (P<0.01) danbagi ladang yang tidak mempunyai doktor veterinar yang tetap (P<0.05).

ABSTRACT

Antimicrobial resistance of 274 E. coli isolates from farm workers, nondiarrhoeic and diarrhoeic piglets to 11antimicrobial agents was investigated to determine the antimicrobial resistance pattern of E. coli. Five pig farmswere involved in this study. Antimicrobial resistance was significantly higher (P<0.01) and almost 100% forsulphasoxazole, streptomycin and tetracyline in porcine isolates compared with human isolates. The meanpercentages of antibiotic resistance between farms were not significantly different (P>0.05) irrespective of sourceof isolate. Statistical analysis showed that the antibiotic resistances were significantly lower (P<0.05) in farmxvorker isolates than in nondiarrhoeic and diarrhoeic piglet isolates. The percentages of isolates resistant to at leasteight antibiotics were 27.7, 5.6 and 1.0 for isolates from diarrhoeic, nondiarrhoeic piglets and farm workersrespectively. The present study also indicated that the mean antimicrobial resistance was significantly higher inlarger farms (P<0.01) and in farms without a resident veterinarian (P<0.05).

INTRODUCTION f o r t h e t r e a t m e n t o f diseases has resulted in the

Frequent and often indiscriminate use of veteri- increase and spread of antimicrobial-resistantnary drugs, particularly the antimicrobial agents, bacteria in animals both overseas (Findland 1975;either in animal feeds as a growth promotant or Dubel et al 1982) and in Malaysia (Khor et al

P Y. CHOO, A. SALAM ABDULLAH AND A.R. CHE NYONYA

1982; Tan and Chin 1982; Bahaman and Liman1985). This endangers both animal and humanhealth because it is generally accepted that re-sistant bacterial strains of animal origin canreach humans in contaminated food products.

Salam et al (1986) have shown that theprevalence of antimicrobial-resistant organismsis correlated with the therapeutic use of antimi-crobial agents in the farm. They found in aninstitutional pig farm that Staphylococcus aureuswas resistant to penicillin and streptomycin (thetwo antibiotics commonly used in this pig farm)as well as letracycline and ampicillin. Otherstudies have shown multiple antibiotic resistancein E. o>/nsolated from pigs (Chin 1983; Bahamanand Liman 1985; Linton 1986).

Although a number of studies abroad haveshown the spread of antimicrobial-resistant or-ganisms, particularly E. coli, from animal to man(Levy et al 1976; Linton et al 1977; O'Brian et al1982; Linton 1986), no such study has beencarried out in Malaysia. Therefore, the objectiveof the present study is to determine the relation-ship between the antimicrobial resistance be-tween E. coli isolated from pigs and pigfarmworkers.

MATERIALS AND METHODS

Five commercial pig farms were selected for thestudy, based on farm size, the availability of aresident veterinarian, and location of the farm.Farm 1 was an institutional farm, with no otherfarms within a 10-km radius. Farms 2 and 5 werethe closest-at least 5 km apart, whereas Farms 3and 4 were located in an intensive pig-farmingarea of 86 farms with about 90,000 pigs. Twofarms (Farms 1 and 5) have residentveterinarians.

The utilization of antibiotics atsubtherapeutic and therapeutic levels in feedand for the treatment of diseases respectively inthese farms over the last five years is shown inTable 1 and Table 2.

All farms used streptomycin, tetracyline, sul-phpnamides with or without trimethoprim com-binations, tylosin, penicillin and chlorampheni-col in the treatment of livestock diseases withinthe last five years. Gentamycin, neomycin andkanamycin were not used in Farm 1.

All categories of livestock feeds were medi-cated. The diets prepared were creep, weaner,grower and breeder rations. Creep feeding wasintroduced to piglets two weeks old. Tetracy-

TABLE 1Antimicrobial drug usage (therapeutic) in pig farms in the last 5 years

Drug Farm 1 Farm 2 Farm 3 Faun 4 Farm 5

Ampicllin

Penicillin

Chloramphenicol

Gen tarn vein

Kanamycin

Neomycin

Streptomycin

Tetracylines

Sulphomides +Trimethoprim

Tvlosin

Yes*

Yes

Yes**

No

No

No

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes Yes Yes

Yes

Yes

*Since 1987

•"•terminated usage 2 years before the study

PERTANIKAJ. TROP. AGRIC. SCI. VOL. 18 NO. 1, 1995

ANTIMICROBIAL RESISTANCE OF E. COLI ISOIATES FROM PIG FARM WORKERS

TABLE 2Antimicrobial drug usage (in feed) in pig farms in the last 5 years

Drug

Neomycin

Tetracyclines

Snip honam ides +trimethoprim

Tylosin/Erythromycin

Penicillin

Lincomycin

Virginiamycin

Farm 1

Yes

Yrs

No

No

No

No

Yes

Farm 2

No

Yes

Yes

No

Yes

No

No

Farm 3

No

Yes

Yes

Yes

Yes

No

No

Farm 4

No

Yes

Yes

Yes

No

Yes

Yes

Farm 5

No

Yes

Yes

Yes

Yes

No

No

cline was used subtherapeutically in all farms.Other antimicrobial agents commonly used atsubtherapeutic level were sulphonamides,penicillins, tylosin, lincomycin and viginiamycin.The antibiotic levels varied between 50-200 ppmin pig feed rations. Except for Farm 1,subtherapeutic feeding was not withdrawn infinisher feeds.

The lactating sows and suckling pigs werehoused in individual farrowing pens with crateson slatted plastic-coated flooring in all farms.Diarrhoeic piglets were normally medicated witha neomycin-kaolin-pectate drench and with anti-microbial therapy peculiar to farm usage.

Sample Collection and TreatmentApproximately 2 g of faeces from five individualnon-diarrhoeic suckling piglets (3-4 weeks old)from different non-diarrhoeic litters were col-lected in presterilized MacCartney bottles. About6-9 ml of stool samples from individual diarrhoeicpiglets in different litters from the non-diarrhoeicspecimens were collected by abdominal squeez-ing and thumping of diarrhoeal contents intosterile test-tubes.

Human faecal samples were collected from4-5 farm volunteers from each farm intopresupplied sterilized autoclave plastic bags.

All samples were kept in ice (4°C) in polvstv-rene ice chests or kept in a refrigerator, transportedto the laboratory and kept at -20*'C until processed.

One gram of a composite faecal samplefrom a respective study group for each farm was

serially diluted in phosphate buffer and platedonto MacConkey agar. Twenty to twenty-five iso-lated colonies growing on agar plates of thehighest dilutions, with 30 to 200 colonies perplate, were randomly selected. These isolateswere picked according to the proportion oflactose and non-lactose fermenters. The cultureswere reisolated in MacConkey agar at 37°C over-night. Single isolated colonies on MacConkeyagar were used in the identification of speciesusing the API-20E Enteric system. Gram stainingwas performed to confirm gram-negative status.E. roli isolates were purified in MacConkey agarand were subcultured in stock culture agar slantsand refrigerated. Data was analysed using analy-sis of variance and comparison of means byKruska-Walls test to determine the significanceof difference between the means.

Antibiotic Susceptibility TestingPure cultures were subjected to antimicrobialsusceptibility testing by the Kirby Bauer tech-nique on Mueller-Hinton agar on duplicate plates.E. co&ATCG strain 25922 was tised as the controlorganism. Antimicrobial susceptibility discs (Difco)were used to determine resistance. The antimi-crobial agents used were Ampicillin-Ap (10|ig),Cephalothin-Cf (30 jLlg), Chloramphenicol-C (30|^g), Neomycin-N (30 |ag), Streptomycin-S (10jag), Tetracycline-Te (30 (ig), Trimethoprim-Tr(5 Hg) and Sulphasoxa/ole-Su (300 j^g).

Zone inhibition diameters were measuredwith vernier callipers (+0.01 mm) and intcpretcd

PERTANIKAJ. TROP. AGR1C SCI. VOL. 18 NO. 1, 1995

P.Y. CHOO, A. SALAM ABDULLAH AND A.R. CHE NYONYA

TABLE 3Incidence of antimicrobial resistance of E. coli in pigs and farm workers

Source of Isolates Number of Isolates/Number Resistant Per cent Resistance

Non-Diarrhoeal Pigs

Farm 1Farm 2Farm 3Farm 4Farm 5

23/2323/2318/1818/1819/19

100100100100100

Overall 101/101 100

Diarrhoeal Pigs

Farm 1Farm 2Farm 3Farm 4Farm 5

15/1517/1719/1918/1819/19

100100100100100

Overall NN s s 100

Faun Workers

Farm 1Farm 2Farm 3Farm 4Farm 5

15/419/320/1319/6

12/8

26.715.865.031.666.7

Overall 85/34 41.2

as resistant, intermediate or sensitive accordingto National Committee for Clinical LaboratoryStandards (NCCLS).

RESULTS

A total of 101, 88 and 85 faecal isolates of E. colifrom nondiarrhoeic and diarrhoeic piglets andfarm workers respectively were tested (Table 3).

All faecal isolates from nondiarrhoeic anddiarrhoeic piglets were resistant to at least oneantimicrobial agent in contrast to 41.2% of iso-lates from pig farm workers. The incidence ofmultiple resistance was observed in all isolatesfrom diarrhoeic piglets and in 96.3 + 2.55% ofisolates from nondiarrhoeic piglets. However,the mean frequency of multiple resistance infarm workers (29.9 + 6.62%) was significantlylower (P<0.01) than in nondiarrhoeic anddiarrhoeic piglets.

In isolates from nondiarrhoeic piglets, theantimicrobial resistance was high for tetracy-cline (98.3%), streptomycin (91.6%) andsulphasoxazole (75.7%) and low for gentamycinand nalidixic acid (0.9%) (Table 4). The resist-ance frequencies were 35.5, 32.3, 31.4 and 24.7%for chloramphenicol, kanamycin, neomycin andampicillin respectively.

In isolates from diarrhoeic piglets, the re-sistance to tetracycline, streptomycin andsulphadimidine was 95.5, 92.3 and 90.9% re-spectively. The resistance to nalidixic acid wasonly 1%. However, the resistance to ampicillin,gentamycin and kanamycin and trimethoprimwas significantly higher (P<0.05) innondiarrhoeic pig isolates (Table 4) even thoughthese antibacterial drugs were not used in feedsin these farms. The resistant bacteria emergedas a result of extensive therapeutic use of these

PERTANIKAJ. TROP. AGRIC SCI. VOL. 18 NO. 1, 1995

ANTIMICROBIAL RESISTANCE OF E. COLI ISOIATES FROM PIG FARM WORKERS

TABLE 4Antimicrobical resistance of isolates of E. colt in non-diarrhoeal

and diarrhoea) pigs and farm workers

Samples

Non-Diarrhoeal

Farm 1Farm 2Farm 3Farm 4Farm 5

Mean

Diarrhoeal Pigs

Farm 1Farm 2Farm 3Farm 4Farm 5

Mean

Farm Workers

Farm 1Farm 2Farm 3Farm 4Farm 5

Mean

AP

Pigs

42627561 1

24.8"

130823733

100

53.0"

75

250

25

12.4-

c

174

8967

0

35.41

2077376768

53.8*

2055

110

8.2'

Antimicrobial

Giti

04000

0.3d

1312110

32

13.6h

00000

0"

K

012

10044

5

32.41

2741478363

52.2*

755

110

5.6*

Agent (%

X

013

10039

6

31.4"

4747428363

56.4"

755

110

5.6"

Resistance)

S

74100100100

84

91.6"

100943483

100

92.2d

1316502616

24.2<

Tfc

91100100100100

98.2d

•10094

1008995

95.6'1

1311503233

27.8*.

Tr

17131728

0

15.0'

47100325695

66.0"

135500

4.61

Su

i>l65

10010063

75.8d

6010010094

100

90.8(i

2716651125

26.8*

a, b Values in different columns bearing different superscripts are significantly difierent (p <0.05).

c, d Values in different columns bearing different superscripts are significantly different (p < 0.01).

antibacterials and colonized the intestine of rion-diarrhoeic piglets. Antimicrobial resistance ofdiarrhoeal isolates to chloramphenicol variedfrom 20 to 76.5% between farms but was notsignificantly different (P>0.05) from thenondiarrhoeic isolates. There was also no signifi-cant difference between the resistance of eitherisolate to streptomycin, tetracycline andsulphasoxazole.

The faecal E. colt isolates from farm workersshowed resistances not exceeding 30% for allantimicrobial agents tested. The resistance ob-served was high for tetracycline (27.7%), strep-tomycin (24.4%) and sulphadimidine (26.6%),Therefore, the antimicrobial resistance in iso-lates from farm workers was similar to isolatesfrom diarrhoeal and non-diarrhoeal porcineorigin although at a significantly lower frequency

(P<G.O1). Resistance to ampicillin and chloram-phenicol was lower, at 12.4% and 8.2% respec-tively. Resistance against cephalothin,gentamycin, neomycin, kanamycin andtrimethoprim was lowest.

The sample sizes of faecal isolates by sourceand farm were too small for conclusive compari-son. However, statistical analysis showed thatantibiotic resitances were significantly lower infarm workers' samples than in i) nondiarrhoeicpiglets for chloramphenicol, neomycin (P<().()5);tetracycline, streptomycin and sulphadiazine(P<0.01) and ii) diarrhoeic isolates for ampicil-lin, chloramphenicol, gentamycin, kanamycin,neomyin (P<0.05); tetracycline, streptomycin,sulphasoxazole and treimethoprim (P<0.01).

However, there was no significant differ-ence (P>0.05) between the variance of mean of

PERTANIKA J. TROP. AGRIC. SCI. VOL. 18 NO. 1, 1995 5

P.Y. CHOO, A. SAIAM ABDULLAH AND A.R. CHE NYONYA

TABLE 5Resistance to number of antibiotics

Per cent Resistant to at Least

Samples

Xon-Diarrhoeai Pigs

Farm 1 87Farm 2 100Farm 3 100Farm 4 100Farm 5 95

1726

1009411

0438939

0

00

11170

Mean I<UVj 34.2* 5.6"

Diarrhoea] Pigs

Farm 1Farm 2Farm 3Farm 4Farm 5

100100100100100

531006389

100

27SL>177290

1329112263

Mean 100.0'1 S I . I ' 1 63.6d 27.6b

Farm Workers

Farm 1Farm 2Farm 3Farm 4Farm 5

\(\552625

13112011o

1355

110

00500

Mean 29.8( L0"

a,b Values in the same column bearing different superscripts are significantly different (P < 0.05).

C, d Values in the same column bearing diferent superscripts are significantly different (P < 0.01).

antimicrobial resistance between farms for allthree groups of isolates.

A total of 27, 33 and 16 antibiotypes wereobserved from nondiarrhoeic pigs, diarrhoeicpigs and farm workers respectively. The morecommon antibiotypes were i) in nondiarrhoeicpig isolates, TeSSu (26.7%), TeS (13%) andTeSSuNKC (10.9%); ii) in diarrhoeic isolates,TeSSuNKCAPTr (17.1%), TeSSuCApTr (13.6%)and TeSSu (9.1%); and iii) in farm workers,TeSSu (71.%) and TeSSuAp (4.7%) were mostcommon.

The median resistance was 3.8, 6.5 and 0.8antibiotics for nondiarrhoeic piglets, diarrhoeicpiglets and farm workers respectively. Table 5shows 27.7% of diarrhoeic porcine isolates wereresistant to at least eight antibiotics in compari-son with 5.6 and 1.0% observed in nondiarrhoeicpig and farm worker isolates (P<0.05) respec-

tively. With respect to resistance to at least fiveantibiotics, 74.3% of diarrhoeic were resistantcompared with 39.9 and 6.8% in nondiarrhoeicpigs and farm workers isolates, respectively.

The mean multiple antimicrobial resistance(MAR) index, which is defined as the meannumber of resistance observed over the numberof antibiotics tested expressed as a percentage(Krupperman 1983), showed that diarrhoeic pigshad the highest MAR index (Table 5).

There were significant differences (P<0.05)in MAR index between the three groups ofisolates. No significant difference in the meanMAR index between farms was observed.

DISCUSSIONIn all five farms tinder study, the frequency ofantimicrobial resistance observed in diarrhoealswine isolates was significantly higher than those

PERTANIKA J. TROP. AGRIC. SCI. VOL. 18 NO. 1, 1995

ANTIMICROBIAL RESISTANCE OF E. COLI ISOIATES FROM PIG FARM WORKERS

of nondiarrhoeic isolates for ampicillin,gentamycin, kanamycin and trimethoprim(P<0.05), although these four antibacterial drugswere not included in feeds. This suggests that theindiscriminate therapeutic use of antimicrobialdrugs is more likely to cause resistance than theuse of these drugs in feeds as a growth promotantOur results, which showed a higher prevalence ofantimicrobial resistance in large herds with noveterinary supervision than in small herds withveterinary supervision, support our hypothesis.This finding also concurs with a study on pigs inan institutional farm (Salam et al 1986) on thepattern of resistance of K coli and Staphylococcusaur€usy which showed increased resistance to drugsin the treatment of livestock in the farm.

On an individual farm basis, Farm 1 hadlower antibiotic resistance frequencies for chlo-ramphenicol, ampicillin and kanamycin. In thisfarm the use of chloramphnicol had been termi-nated two years earlier. Ampicillin was recentlyintroduced in the treatment of piglet diarrhoeawhile kanamycin was not used. However, in Farm5 where ampicillin was the therapeutic drug ofchoice, all diarrhoeic strains isolated were resist-ant to it. Individual farm preference for the useof chloramphenicol (Farm 2) and kanamycin(Farm 4) for therapeutic use was associated withhigher antibiotic resistance frequencies for thesedrugs in porcine isolates.

The observation of antibiotic resistance pat-terns of human isolates cannot be comparedwith piglets' isolates directly as they may not beepidemiologically linked. An extensive epide-miological survey is therefore required involvinga larger number of farms. However, theantibiogram characteristics of faecal isolates ofE. coli from farm workers is significantly lowerthan nondiarrhoeic piglet isolates for chloram-phenicol and neomycin (P<0.05), streptomycin,tetracycline and sulphadimidine (P<0.01). It is,therefore, consistent with the finding that mostantibiotic resistance in human pathogens relatesto the use of antimicrobial drugs in human,rather than in veterinary medicine (Prescott andBaggot 1988). Our results also indicated that thebacteria may develop resistance and colonizehuman intestines irrespective of whether thedrugs are used therapeutically or prophylacticallyin farms. Thus, it is unfair to ban or withdrawthe antibiotic usage in animal feeds as a growth-promotant while allowing the indiscriminate useof these drugs therapeutically.

The farm workers involved in our study didnot report increased incidences of diarrhoealepisodes while working with livestock. This agreeswith the findings of Linton et al. (1977), thatthere is little evidence that strains of E. colienteropathogenic for man arise from animals.

Although no significant differences (P>0.05)were observed between the mean resistances toantibiotics tested between farms for all threegroups of isolates, it was observed that there wasa trend towards higher incidence of antimicro-bial resistance with increasing herd size (Farm 1versus Farms 2-5). Lower antimicrobial resist-ance prevalence was observed in nondiarrhoeicswine isolates and in farms with veterinary super-vision (Farm 1 and 5 versus Farms 2, 3 and 4).However, statistical analysis was not performedas the sample sizes were too small for compari-son, and this observation must be substantiatedby more extensive epidemiological studies. An-other observation which tended to agree withthe findings of Bahaman and Liman (1985) wasthat multiple antimicrobial resistance to thenumber of antibiotics was higher in large thanin small farms. This may explain the differencesobserved between Farms 1 and 5 (Farm 1 had 50sows versus Farm 5 which had 500 sows).

ACKNOWLEDGEMENTS

The authors wish to thank the Director Generalof Veterinary Services for his support in publish-ing this paper. The work is part of the seniorauthor's MSc research project at UniversitiPertanian Malaysia.

REFERENCESBAHAMAN, A.R. and B. LIMAN. 1985. The prevalence

of antimicrobial resistance in £ coli in pigsfrom Johor and Selangor. Malaysian AppliedBiology 14: 19-23.

CHIN, H.C. 1983. Colibacillosis in piglets, the dis-ease and its prevention. Malaysian LivestockCooperative Magazine No. 1.

DI:BEL, J.R., D.L. ZINK, L.M. KKI.LY, S.A. NAQI, H.M.

RENSHAW. 1982. Bacterial antibiotic resistance:Frequency of gentamicin-resistant strains ofEscherichia coli in the faecal microflora of com-mercial turkeys. American Journal of VeterinaryResearch 43: 1786-1790.

FiMii \\[>, M. 1975. Relationship of antibiotics inanimal feeds and salmonellosis in animals andman. Journal of Animal Science 40: 1222-1236.

PERTANIKAJ.TROP.AGRIC. SCI. VOL. 18 NO. 1, 1995

P.Y. CHOO, A. SALAM ABDULLAH AND A.R. CHE NYONYA

KHOR, M., Y.S. LIM and M. JEGATHESEN. 1982. Antibi-

otic resistance and R plasmids in coliformsisolated from some Malaysian cooked food. InAbstracts of 5th Malaysian Microbiology Sympo-sium, Kuala Lumpur,

KRUPPERMAN, P.H. 1983. Multiple antibiotic resist-ance indexing of E. coli to identify7 high risksources of faecal contamination of foods. Ap-plied Environmental Microbiology 46: 165-170.

LEVY, B., G.B. FITZGERAJLP, A.B. M M N >M•:. 1976. Spread

of antibiotic resistant plasmids from chickento chicken and from chicken to man. Nature260: 40-42.

LINTON, A.H. 1986, Flow of resistance genes in theenvironment from animals to man. Journal ofAntibiotic Chemotherapy. Supp.C. 189-197.

LINTON, A.H., K. HOWE, P.H. BENNETT, H.M. Ricnard

and EJ. WHITESIDE. 1977. The colonization ofthe human gut by antibiotic resistant B. rollfrom chickens. Journal of Applied Bacteriology 43:465-469,

O'BRIAN, T.F., J.D. HOPKINS, E.S. GII.LEECE, A.A.

MKDERIOES, K,L. KENT, H.O. BLACHBURN, M.H.

HOLMES, J.P. PEARDON, J.M. VERGERONT, W.L.

SCHELL, E. CHERISTENSON, M.L. BISSKJT and E.V.

MORSE. 1982. Molecular epidemiology of anti-biotic resistance in salmonella from animalsand human beings in the United States. NewEngland Journal of Medicine 307: 1-6.

PRESCOTT, J.F. and J.D. BAC.GOT. 1988. Antimicrobial

Therapy in Veterinary Medicine. Boston: Blackwell.

SAIAM ABDUUAH, A., J.D. BtApGOT and H.L. Too.

1986. Antimicrobial therapy and bacterial re-sistance: A pilot survey. Malaysian Applied Biol-ogy 15: 185-185.

TAN, E.L. and H.C. CHIN. 1982. Antibiotic sensitivityof gram negative bacteria of porcine origin. InAbstracts of 5th Malaysian Microbiology Sympo

- sium, Kuala Lumpur.

(Received 12 June 1991)(Resubmitted 10 January 1995)

PERTANIKAJ. TROP. AGRIC. SCI. VOL. 18 NO. 1, 1995