当前位置：文档库 › Detection of Antimicrobial Substances in Individual Cow

Detection of Antimicrobial Substances in Individual Cow

1999J Dairy Sci 82:704–711

704

Received July 16,1998.

Accepted December 8,1998.

1To whom correspondence should be addressed.Detection of Antimicrobial Substances in Individual Cow and Quarter Milk Samples Using Delvotest Microbial Inhibitor Tests

J.ERIC HILLERTON,*,1BEN I.HALLEY,?PAUL NEAVES,?

and MARTIN D.ROSE §

*Institute for Animal Health,Compton,Newbury,

RG207NN,United Kingdom

?University of Vermont,Burlington,VT 55405

?Williams &Neaves,Moleview,

28Randalls Road,Leatherhead KT227TQ United Kingdom

§CSL Food Science Laboratory,Ministry of Agriculture,Fisheries &Food,

Norwich Research Park,Colney,Norwich NR47UQ,United Kingdom

ABSTRACT

The use of antibiotic therapy to treat and prevent udder infections of cows during the dry period is a key component of mastitis control in many countries.At the same time,the general public is becoming in-creasingly aware of potential hazards from antibiotic residues in foods.Consequently,Delvotest Cow Test (Royal Gist-brocades NV,Delft,The Netherlands),an on-farm version of Delvotest P,a microbial inhibi-tor test for antimicrobials,is being increasingly used by farmers to assess that milk from individual cows is fit for consignment to the bulk tank.Occasional reports of unexplained positive test results have led to suggestions of possible false-positive reactions in milk from individual cows.To investigate the potential causes of such positive test results,three separate investigations were undertaken.In a field survey of unexplained positive reports from farmers,14milk samples from six farms that tested positive were all found to contain antibiotic residues.In more formal investigations of individual quarter milk samples from an experimental herd,none of 134milk samples from midlactation cows yielded positive reactions;for cows that had just calved,16of 144milk samples were positive,and,of those,13had somatic cell counts >4,000,000/ml.Natural inhibitors were respon-sible for 1positive reaction,8positive reactions were related to incomplete milking,and 7samples con-tained b -lactam antibiotics.Positive reactions caused by antibiotic persisted in individual quarter samples for up to 7d postcalving compared with 4d for milk samples from the whole udder.Delvotest was sensi-tive to cephalonium,the active ingredient of Cepravin

Dry Cow (Mallinckrodt Veterinary Ltd.,Uxbridge,United Kingdom),which is the market-leading product in the United Kingdom.Test results yielded a partial purple color reaction in the presence of 8m g/kg of cephalonium and a completely purple reaction at 16m g/kg.These results confirm the validity of Del-votest when used to examine composite milk samples from individual cows supplying the United Kingdom dairy industry and suggest that,with proper atten-tion to milk withdrawal periods and complete milk-ing,there is no obvious risk of antibiotic contamina-tion of milk.

(Key words :b -lactam antibiotics,raw milk,microbial inhibitor test,somatic cells)

Abbreviation key :HACCP =Hazard Analysis Crit-ical Control Point.

INTRODUCTION

The general public is becoming increasingly aware of food safety issues and the potential for chemical and microbiological hazards in https://www.wendangku.net/doc/8f16018720.html,k is per-ceived by the consumer to be natural and wholesome,and many countries impose severe financial penalties on farmers and,in some cases,veterinary surgeons who allow bulk tank milk to become contaminated with detectable levels of antimicrobials.

Treatment of cows during the dry period with a long-acting formulation of intramammary antibiotic is a key component of mastitis control.In addition,the dairy industry in the United Kingdom is seeking to apply the Hazard Analysis Critical Control Point (HACCP )system to milk production;verification is one of the seven HACCP principles.Consequently,there is a need for an on-farm test for antimicrobials so that farmers can ensure that milk from individual cows is fit for consignment to the bulk tank.Delvotest Cow Test (Royal Gist-brocades NV,Delft,The

DETECTION OF ANTIMICROBIALS IN MILK FROM COWS705

Netherlands)is the only screening test available to confirm that milk from an individual cow,especially milk from freshly calved cows that have been treated with a dry cow therapy,may be consigned to the bulk tank.This test is an easy-to-use version of Delvotest P(ampule version;Royal Gist-brocades NV)and is used as the industry standard test in the United Kingdom and elsewhere for the assessment of bulk tank milk.Delvotest Cow Test is based on the microbial inhibition principle and can be considered interchangeable with Delvotest P.

A positive test result does not identify an anti-microbial substance but indicates that the milk is abnormal in some way.Positive test results are rela-tively uncommon,but simple laboratory confirmatory tests can demonstrate whether or not the microbial inhibitor is a b-lactam compound.However,confirma-tory tests are not always used,and occasional reports of positive Delvotest results for milk from individual cows,sometimes for prolonged periods after calving, have led some veterinary pharmaceutical companies to suggest the possibility of false-positive reactions. These reactions have been purported to be associated with high SCC or the presence of the natural inhibi-tors(i.e.,lysozyme and lactoferrin)(1).In addition, an evaluation of the use of antibiotic tests for in-dividual cow samples concluded that the presence of natural inhibitory substances may give rise to unac-ceptable rates of false-positive reactions(5).To in-vestigate potential causes of such positive test results and to determine whether the test could provide valu-able information regarding milk samples from in-dividual cows,a series of three separate investiga-tions was undertaken.

In an initial field survey,farmers who reported unexplained positive Delvotest results for milk from individual cows submitted samples to an independent laboratory to identify the nature of the inhibitory substances.The objective was to determine whether or not samples contained an antibiotic and,if possi-ble,to characterize the substance present.

To investigate whether any relationship existed between the presence of antimicrobial substances and high SCC in milk samples from individual cows and to determine whether the composition of milk produced at different times during the lactation cycle affected the test result,a more comprehensive investi-gation was undertaken using much greater numbers of samples from an experimental herd whose ex-posure to antimicrobial substances was well con-trolled.

During the field survey,it became apparent that one proprietary dry cow therapy,Cepravin Dry Cow (Mallinckrodt Veterinary Ltd.,Uxbridge,United Kingdom),was strongly associated with the majority of positive test results.The active ingredient of Cepravin is cephalonium,a cephalosporin for which no minimum residue limit has yet been set in Euro-pean Union legislation.To assist farmers,veterinary surgeons,and others involved with milk production in assessing the significance of contamination of the milk supply with cephalonium,the sensitivity of Del-votest to this antibiotic was determined.

This paper reports the results of all three investi-gations and recommends usage methods for Delvotest to ensure the validity of test results.

MATERIALS AND METHODS

Field Survey

Test samples.During1996and1997,14samples were collected from six farms that had experienced positive results using Delvotest.Twelve of these sam-ples were composite samples from individual cows, and2were samples from bulk tank milk.All farmers completed a written questionnaire that provided in-formation regarding antibiotic treatment of the herd in general and information regarding the problem cow in particular.No attempt was made to bias the sam-ple toward the use of any particular dry cow prepara-tion.All notifications of on-farm positive samples were investigated.

All samples were transported overnight under chilled conditions and then examined at the Leather-head Food Research Association Laboratory (Leatherhead,United Kingdom)(the former Microbiology Laboratory of the Milk Marketing Board for England and Wales),which held NAMAS(Na-tional Measurements Accreditation Scheme)accredi-tation for the detection of antibiotics in milk using Delvotest.In addition,1negative individual cow sam-ple was tested without the laboratory’s prior knowledge of the expected result to confirm that a truly negative milk sample would yield negative results under all of the identification tests.The sam-ple was collected from a cow that had no recent history of antibiotic treatment.

Test methods.All samples were examined using 1)Delvotest P(ampule format),a microbial inhibitor test that detects antimicrobial substances nonspecifi-cally;2)Delvo X-Press b-L(Royal Gist-brocades NV),an immune receptor test that specifically de-tects b-lactam antibiotics;3)SNAP test for b-lactams (IDEXX Laboratories Ltd.,Chalfont St.Peter,United Kingdom),an immune receptor test that specifically detects b-lactam antibiotics,and4)LacTek b-lactam test(IDEXX Laboratories Ltd.),an immune-receptor

Journal of Dairy Science Vol.82,No.4,1999

HILLERTON ET AL. 706

test that specifically detects penicillins but does not detect cephalosporins.

Interpretation of test results.A positive reaction to Delvotest indicates the presence of any substance that prevents the growth of the test organism.The test was considered positive when a complete purple reaction was observed(i.e.,any partial color changes were considered negative).This interpretation is in accordance with the Code of Practice for the Assess-ment of Milk Quality(4),which provides a uniform basis for raw milk testing throughout England and Wales.

A positive reaction in any of the immune receptor tests unequivocally indicated the presence of a b-lactam antibiotic.However,because these tests differed in the range of b-lactams they could detect,a sample that passed the LacTek test but failed the SNAP or Delvo X-Press test was considered to contain a cephalosporin.

Confirmation techniques.To confirm the presence of a b-lactam,all positive samples were retested after treatment with penicillinase(penase), an enzyme that destroys b-lactam antibiotics.A nega-tive reaction after penase treatment confirmed the presence of a b-lactam antibiotic.

To confirm the presence of a natural inhibitor,all positive samples were retested after heat treatment at82°C for5min,a treatment that destroys lysozyme and lactoferrin.A negative reaction after heat treat-ment confirmed the presence of a natural inhibitor. Investigation of Milk

from an Experimental Herd

Test samples.Four hundred forty-three milk sam-ples were obtained for three trials from the Holstein cows at the Institute for Animal Health(Compton, United Kingdom).Complete health and medication records were available for each cow.No cow had received any antimicrobial treatment within28d of sampling.All cows that had recently calved had received a cephalonium intramammary infusion (Cepravin Dry Cow)as part of drying off treatment at least8wk before calving.

For the first investigation,midlactation cows likely to have SCC in the range from<400,000to >4,000,000/ml in the composite milk sample were selected after inspecting the most recent monthly in-dividual cow SCC results.In the second investigation, samples were taken from cows at4to6d postpartum when milk was first entering the bulk tank.The SCC results in the second investigation were,on average, high,and50%of the samples had>1,000,000cells/ml in the absence of infection,suggesting incomplete milking.Therefore,a time series of samples was ob-tained from10cows that had recently calved that were confirmed by inspection to be properly milked out.Sampling began<24h postcalving and continued for10d postpartum or until no evidence of an-timicrobial activity was found in any sample for2 consecutive d.

Milk samples were collected each morning as in-dividual quarter foremilk samples or as whole udder composite samples from the recorder jar of the milk-ing machine.Quarter samples(20ml)were obtained routinely by the International Dairy Federation(8) method.The composite samples were obtained by agitating the whole milk for5s by air admission and then collecting20ml through the drain valve.

Test methods,interpretation of results,and confirmation techniques.All samples were exam-ined the same day for the presence of inhibitory sub-stances using Delvotest Cow Test.The instructions provided with the test kits were followed exactly. Samples were incubated for exactly 2.5h at64±0.5°C,and milk was considered to contain inhibitory substances if the entire vial contents remained pur-ple.

To confirm the presence of a natural inhibitor,an aliquot of any sample containing inhibitory sub-stances was retested after heat treatment at82°C for 5min.To confirm the presence of a b-lactam antibi-otic,a0.25-ml aliquot of any sample still exhibiting inhibitory activity after heat treatment was incubated with1000U of penicillinase(Cilinase;Consolidated Chemicals,Wrexham,United Kingdom)for15min and retested.

Milk SCC were determined on aliquots of the sam-ples using the IDF Coulter counter method(8).The electrical conductivities of samples taken in the time series study were determined using an Eisai Milk-Checker(7).

Quality control samples.Quality control samples containing antibiotic-free milk or milk containing 0.004,0.006,or0.008IU of penicillin G/ml or10or14 m g of cephalonium/kg were supplied by the United Kingdom Milk National Reference Laboratory(CSL Food Science Laboratory,Norwich,United Kingdom). These samples were stored at4°C and were included randomly on the testing occasions.

Sensitivity to Cephalonium

Preparation of test and control samples.Raw milk samples collected from farm bulk tanks were supplied by the Milk Marque Quality Testing Labora-tory(Milk Marque,Newcastle-under Lyme,United Kingdom).All milk had passed the Milk Marque

Journal of Dairy Science Vol.82,No.4,1999

DETECTION OF ANTIMICROBIALS IN MILK FROM COWS

707

TABLE 1.Identity of antibiotics in milk samples that had failed an on-farm Delvotest (Royal Gist-brocades NV,Delft,The Netherlands).

1According to farmer.

2IDEXX

Laboratories Ltd.(Chalfont St.Peter,United Kingdom).

3Royal Gist-brocades NV.

4Mallinckrodt Veterinary Ltd.(Uxbridge,United Kingdom).5F =Fail,P =pass,NT =not tested.

6Solvay-Duphar Veterinary Ltd.(Southampton,United Kindgom).

Laboratory examination Untreated

Dry cow

SNAP Delvo Milk sample

therapy used 1Delvotest Test 2X-Press 3LacTek 2Penase Heat Conclusion Farm 1Cow A Cepravin Dry Cow 4F 5F F P P F Cephalosporin Farm 2Cow A Cepravin Dry Cow F F F P P F Cephalosporin Farm 3Bulk tank Cepravin Dry Cow F F F F P F Penicillin Farm 4

Cow A Kloxerate Plus DC 6F F F F P F Penicillin Cow A Kloxerate Plus DC F F F F P F Penicillin Cow A Kloxerate Plus DC F F F F P F Penicillin Cow B Kloxerate Plus DC F F F F P F Penicillin Cow C

Kloxerate Plus DC F F F F P F Penicillin

Farm 5Cow A

Cepravin Dry Cow P F F P P F Cephalosporin Cow A Cepravin Dry Cow P F F P P F Cephalosporin Cow B Cepravin Dry Cow F F F P P F Cephalosporin Cow B Cepravin Dry Cow P F F P P F Cephalosporin Cow C

Cepravin Dry Cow P F F P P F Cephalosporin Farm 6Bulk tank Cepravin Dry Cow F F F F P F Penicillin Negative control

None

No antibiotic

screening test for antibiotics.All samples (50×30ml)were pooled into a single batch for the study.A stock solution of cephalonium (Mallinckrodt Veterinary Ltd.)was prepared in 20%(vol/vol)aque-ous ethanol.Working dilutions (0to 15m g/ml)were prepared from the stock solution in 0.2M phosphate buffer (pH 6.0).Aliquots of the pooled milk were augmented at 0,2,5,8,10,12,14,16,18,20,22,and 25m g/kg by adding 100m l of working dilution to 49.9g of milk.Six replicates were examined from the augmented milk samples.

Negative controls were aliquots from the pooled batch of milk to which no additions had been made and aliquots (49.9g)containing phosphate buffer plus 4%(vol/vol)ethanol (100m l).Positive controls were aliquots from the pooled batch of milk contain-ing penicillin G (0.004,0.006,and 0.008IU/ml)or cloxacillin (20,25,and 30m g/kg).All controls were examined in triplicate.

Examination of samples.All samples were exa-mined using Delvotest P (ampule format).All tests were undertaken according to the instructions of the manufacturer.The ampules were incubated at 64±0.5°C in a heating block for up to 2.5h.Color reac-tions were observed visually at six 5-min time inter-vals from 2to 2.5h by removing the ampules from the

heating block for <2s.Color reactions were recorded using the following notation:yellow throughout the agar phase,YYY;mainly yellow with some purple,YYP;some yellow but mainly purple,YPP;purple throughout the agar phase,PPP.

Milk samples were examined on two separate occa-sions using the same batch of pooled milk (on the second occasion,after storage at 5°C for 24h)and a different batch code of Delvotest P.Solutions of cephalonium,penicillin G,and cloxacillin were pre-pared separately on each occasion.

RESULTS AND DISCUSSION

Field Survey

The field survey results are shown in Tables 1and 2.b -Lactam antibiotics were detected in all test sam-ples,and the negative control sample yielded negative results for all tests.All samples that had yielded positive Delvotest results when examined by the farmer were thus confirmed as true positives.Positive individual cow samples were associated with the use of two dry cow therapies,Cepravin Dry Cow,which contains the cephalosporin cephalonium,and Kloxer-ate Plus DC (Solvay-Duphar Veterinary Ltd.,

Journal of Dairy Science Vol.82,No.4,1999

HILLERTON ET AL.

708

TABLE 2.Antibiotic treatment history of individual cows producing milk contaminated with antibiotics.1According

to farmer.

2Mallinckrodt Veterinary Ltd.(Uxbridge,United Kingdom).3Not reported.

4Individual cow samples that fulfilled the withholding period criteria for inclusion in the bulk tank.5Solvay-Duphar Veterinary Ltd.(Southampton,United Kindgom).

Dry cow

Dry Time after Withdrawal Time beyond

Milk sample therapy used 1period calving period withdrawal period (d )(d )

Farm 1Cow A Cepravin Dry Cow 2NR 3NR 51d +96h 71,postinfusion Farm 2Cow A 4Cepravin Dry Cow 224651d +96h 13Farm 4

Cow A Kloxerate Plus DC 532030d +96h –4Cow A Kloxerate Plus DC 32230d +96h –2Cow A 4Kloxerate Plus DC 321230d +96h 10Cow B 4Kloxerate Plus DC 261830d +96h 10Cow C Kloxerate Plus DC 41330d +96h –1Farm 5

Cow A 4Cepravin Dry Cow 50951d +96h 4Cow A 4Cepravin Dry Cow 501251d +96h 7Cow B Cepravin Dry Cow 53451d +96h 2Cow B 4Cepravin Dry Cow 53651d +96h 4Cow

Cepravin Dry Cow

d +96

–3

TABLE 3.Relationship between SCC and Delvotest (Royal Gist-brocades NV,Delft,The Netherlands)reaction in individual quarter milk samples from midlactation cows.

Delvotest reaction

SCC Sample

Negative Positive

(no.)

(1000/ml)<400

1101100400–1000161601001–4000660>4000220Total

134

0Southampton,United Kindgom),which contains the penicillins cloxacillin and ampicillin.All samples from individual cows associated with the use of Cepravin passed the LacTek test and were thus consi-dered to contain a cephalosporin;however,Cepravin was not responsible for either of the two bulk tank failures.

In contrast to the results of Cullor et al.(5),Delvotest yielded no false-positive reactions (i.e.,all positive samples contained a b -lactam compound,and natural inhibitors were not detected in any sample).Possible explanations for these contrasting results can be suggested;however,detailed comparisons be-tween the two studies are difficult because,in the study by Cullor et al.(5)positive samples were not examined for the presence of b -lactam antibiotics and the proportion of purple color in the test ampule that those workers considered positive was not reported.Investigation of Milk

from an Experimental Herd

All positive and negative control samples yielded the expected reactions.

Examination of 134individual quarter samples col-lected from midlactation cows yielded only negative results;no inhibitory substances were detected (Ta-ble 3).Twenty-four samples had SCC >400,000/ml,which suggests that no obvious correlation exists be-tween SCC in milk from midlactation cows and inhi-bition of the test organism.

Sixteen of 144individual quarter samples collected from cows at 4to 6d postcalving yielded positive Delvotest reactions (Table 4).Thirteen of these were associated with SCC >4,000,000/ml.Thus,for early lactation cows,it appeared more likely that high SCC milk would contain inhibitory substances,especially at an SCC >4,000,000/ml.Heat treatment rendered only 1of the 16samples negative (SCC >4,000,000/ml),suggesting a low incidence of natural inhibitors.A further 7samples yielded negative results after penase treatment,indicating that they contained a b -lactam antibiotic.Because of the antibiotic history of the cows used in the experiment,the presence of Cepravin seemed likely.The remaining 8positive samples (i.e.,one-half of those that were positive for inhibitory substances in the initial screening)con-

Journal of Dairy Science Vol.82,No.4,1999

DETECTION OF ANTIMICROBIALS IN MILK FROM COWS

709

TABLE 5.Detection of inhibitory substances in quarter and whole udder milk samples from early lactation cows using Delvotest (Royal Gist-brocades NV,Delft,The Netherlands).

Whole udder samples

Quarter samples

Positive reaction

Positive reaction Milk After heat Milk After heat samples Initial After heat and penase samples Initial After heat and penase Cows

Postpartum examined test treatment treatment examined

test

treatment

represented

(d )(no.)1102204017143621022040101004322201799034221010770352000977036200075403700007210280000500009

TABLE 4.Relationship between SCC and Delvotest (Royal Gist-brocades NV,Delft,The Netherlands)reaction in individual quarter milk samples from cows that recently calved.

Delvotest reaction

SCC Sample

Negative Positive

(no.)

(1000/ml)<400

42402400–1000282711001–400032320>4000422913Total

144

128

16tained inhibitors that were resistant to both penase and heat treatment.All samples had SCC >4,000,000/ml,which was considered abnormally high for unin-fected quarters at this stage postpartum.Farm obser-vations revealed that lactation had been improperly started by a period of incomplete milking.The inhibi-tory substances were not identified.The results sug-gest the potential for inhibitory substances to occur in quarter milk samples with high SCC obtained from early lactation,but the milking technique may be a contributing factor and might have led to an inap-propriate and incomplete milk withdrawal period for the dry cow therapy used.The high SCC also reveals poor establishment of lactation,which may be related to the presence of heat-stable inhibitors in the milk.Examination of quarter milk samples collected over a time series showed that >40%of milk samples collected within 24h postcalving contained inhibitory substances.Most of the samples contained substances that were heat stable,and almost all were sensitive to penase,suggesting that the samples contained b -

lactam antibiotics (Table 5).The nonantibiotic in-hibitory substances disappeared rapidly from quarter milk as lactation was established.However,a penase-sensitive substance persisted in individual quarters of a few cows for up to 7d postcalving.This substance was detected in quarter foremilk samples,and sam-ples containing inhibitory substances came only from certain individual cows.In whole udder milk,all posi-tive samples except 1were due to a b -lactam antibi-otic,but inhibitory substances were never detected in whole udder milk samples beyond the 4-d statutory milk withdrawal period.

Delvotest microbial inhibitor tests can be inhibited by lysozyme or lactoferrin separately and synergisti-cally (2).These inhibitors occur frequently in high SCC milk and are heat labile.Other nonproteina-ceous inhibitors may be present in early lactation and high SCC milk.Samples may contain high concentra-tions of free fatty acids,which may inhibit Delvotest microbial inhibitor tests (3),but the diapedesis of neutrophils may lead to a leakage of serum compo-nents across the mammary epithelium,which,in turn,could lead to lipolysis and inhibitory action (3)and also to an increase in the electrical conductivity of the milk because of an increase in sodium content.The latter is shown to be possible because the electri-cal conductivity of milk from the initial day of the time series,when heat-labile inhibitory substances were present,averaged 7.2mS/cm;the mean for quarter milk from the same cow not containing in-hibitory substances was 5.9mS/cm.Sensitivity to Cephalonium

On both test occasions,all positive and negative control samples yielded the expected reactions.The

Journal of Dairy Science Vol.82,No.4,1999

HILLERTON ET AL.

710

TABLE 6.Sensitivity of Delvotest (Royal Gist-brocades NV,Delft,The Netherlands)to cephalonium.

1Color

reaction after 2.5h at 64°C:YYY =yellow throughout the agar phase,YYP =mainly yellow

with some purple,YPP =some yellow but mainly purple,and PPP =purple throughout the agar phase.2No color change.

Test 1

Test 2

Color Color Color change

Color change

Concentration reaction 1Replicates commenced reaction Replicates commenced (m g/kg)(no.)(no.)0YYY 62h 10min YYY 62h 15min 5YYY 62h 15min YYY 62h 15min 8YYY 32h 15min YYP 62h 15min 8YYP 32h 20min .........10YYP 32h 20min YYP 32h 20min 10YPP 32h 20min YPP 32h 20min 12YPP 62h 20

min

YYY 12h 20min 12.........YYP 12h 20min 12.........YPP 12h 20min

12.........PPP 1NC 214YPP 32h 25min PPP 3NC 14PPP 32h 25min

PPP 6NC 16PPP 6NC PPP 6NC 18PPP 6NC PPP 6NC 20PPP 6NC PPP 6NC 22PPP 6NC PPP 6NC 25PPP 6NC PPP 6NC 30

PPP

test sensitivity for samples containing penicillin G or cloxacillin was in accordance with manufacturer claims,and the addition of phosphate buffer plus ethanol to negative milk samples showed no inhibi-tory effect.

In the presence of cephalonium,Delvotest showed consistent reactions,both among the six replicates examined on the same occasion and between the two test occasions.The combined results given in Table 6show that 12of 12ampules showed a YYY reaction at 5m g of cephalonium/kg,and 9of 12showed YYP at 8m g/kg.A PPP reaction was shown by 12of 12ampules at 16m g/kg.The Code of Practice for the Assessment of Milk Quality (4)defines a PPP reaction as a failure,and all other reactions as pass.On this basis,Del-votest can detect 16m g of cephalonium/kg when added to milk as a pure solution,but no maximum residue limit for cephalonium in milk has been set to date in the European Union.The European Union limit for antibiotic contamination of milk is equivalent to 4m g of penicillin G/kg.

CONCLUSIONS

The field survey provided no evidence of false-positive results when Delvotest was used for milk samples from individual cows.No evidence existed that positive results were due to errors in the test procedure.The test,therefore,appeared to be suffi-ciently rugged for on-farm use,and the widespread belief that the test produces false-positive reactions for individual cow milk samples was not substan-tiated.Indeed,difficulty was encountered in obtain-ing positive samples,which suggests that the problem may not be as widespread as often claimed.

In the more detailed investigations,there was no evidence of positive Delvotest reactions from natural inhibitors in individual quarter milk samples from midlactation cows.Positive Delvotest reactions may occur in individual quarter samples from freshly calved cows because of inhibitory substances other than antibiotics.However,such positive reactions were infrequent and were strongly associated with high SCC (>4,000,000/ml).The results are in close agreement with those reported by Halbert et al.(6)who failed to find natural inhibitors in quarter milks from 407lactating dairy cows and with Gist-brocade data (R.Beukers,1998,personal communication).There was no evidence of natural inhibitors in compo-site milk samples collected from individual cows >4d after calving.Samples collected in accordance with the Delvotest testing protocol are,therefore,unlikely to contain natural inhibitors.To minimize further the occurrence of natural inhibitors in milk from cows

DETECTION OF ANTIMICROBIALS IN MILK FROM COWS711

that recently calved,it is important to adhere to correct milking procedures.

The investigation of the sensitivity of Delvotest to cephalonium has demonstrated that the test sensitiv-ity was in agreement with the manufacturer claim of 15to20m g/kg.The failure of the field survey to detect any bulk tank failures from Cepravin may,therefore, be due to the presence of only low concentrations of cephalonium in individual cow milk samples that are diluted when the milk is combined in the bulk tank.

The results from these investigations confirm the validity of Delvotest when used to examine composite milk samples from individual animals on commercial farms in the United Kingdom.Delvotest,therefore, provides an effective means of independent verifica-tion that is one of the principles of HACCP currently being applied to farms to ensure the safety of the milk supply in the United Kingdom.

We recommend that full withholding times for milk be respected for cows that have recently calved,that only whole udder samples be tested,and that com-plete milk out is achieved.Only if a positive test result is found should quarter samples be inves-tigated,and this should include the use of the Califor-nia Mastitis Test or a measurement of electrical con-ductivity of the milk.

ACKNOWLEDGMENTS

The authors are indebted to those farmers who kindly provided milk samples and cow treatment data for the field survey and to Peter Hopwood(Millwood Products,Bath,United Kingdom)for close liaison with the farmers during the investigation.The help-ful assistance of Quentin Smith(Milk Marque Qual-ity Testing Laboratory,Newcastle-under Lyme, United Kingdom)in providing raw milk samples is gratefully acknowledged.The authors also express their sincere thanks to John Bygrave(CSL,Norwich, United Kingdom)who undertook the practical work for the Delvotest sensitivity studies.

REFERENCES

1Carlsson,A?.1984.Inhibitory substances in milk.M.S.Thesis, Swed.Univ.Agric.,Uppsala,Sweden.

2Carlsson,A?.,and L.Bjo¨rk.1987.The effect of some indigenous antibacterial factors in milk on growth of Bacillus stearother-mophilus https://www.wendangku.net/doc/8f16018720.html,chwissenschaft42:282–285.

3Carlsson,A?.,and L.Bjo¨rk.1992.Liquid chromatography verifi-cation of tetracycline residues in milk and influence of milk fat lipolysis on the detection of antibiotic residues by microbial assays and the Charm II test.J.Food Prot.55:374–378.

4Code of Practice for the Assessment of Milk Quality(1985).

1994.Section7A and7B.United Kingdom Joint https://www.wendangku.net/doc/8f16018720.html,k Marketing Board and Dairy Trade Fed.,London,United King-dom.

5Cullor,J.S., A.van Eenennaam,J.Dellinger,L.Perani, W.Smith,and L.Jensen.1992.Antibiotic residue assays:can they be used to test milk from individual cows?Vet.Med.87: 477–494.

6Halbert,L.W.,R.J.Erskine,P.C.Bartlett,and L.J.Gilbert II.

1996.Incidence of false-positive results for assays used to de-tect antibiotics in milk.J.Food Prot.59:886–888.

7Hillerton,J.E.,and A.W.Walton.1991.Identification of sub-clinical mastitis with a hand-held conductivity meter.Vet.Rec.

128:513–515.

8International Dairy https://www.wendangku.net/doc/8f16018720.html,boratory methods for use in mastitis work.IDF Bull.132,IDF,Brussels,Belgium.

Journal of Dairy Science Vol.82,No.4,1999