antibiotic therapeutics in laboratory animals

REVIEW ARTICLE

Antibiotic therapeutics in laboratory animals

T. H. MorrisDepartment of Laboratory Animal Science, Smith Kline Beecham Pharmaceuticals, New Frontiers SciencePark, Third Avenue, Harlow, Essex CM19 SAW, UK

SummaryInformation on antibiotic therapeutics in laboratory species, especially in rodents andrabbits, is reviewed. A number of areas are considered: interference by antibiotics with anexperiment, antibiotic toxicity, routes of administration, effects of formulation onbioavailability, antibiotic prophylaxis, use of combinations of antibiotics, misuse ofantibiotics, regulatory approval for antibiotic use in animals, sources of information onantibiotic indications and dose, and extrapolation of dose information from other species.

Keywords Antibiotics; laboratory animals; therapy

The purpose of this review is to highlightsome specific questions and problemsconcerning the use of antibiotics in thecommon laboratory species. Although dogs,cats, horses and domestic species are usedin research, they will not be discussed hereas there is a large volume of informationon the clinical use of antibiotics in thesespecies, and as many of these antibioticsare approved by regulatory authorities foruse in these species, clear indications fortheir rational use are provided (NationalOffice of Animal Health 1992, VPD 1991).

This review deals principally withrodents and rabbits since they make upover 90% of the animals used in research.There is no uniform source of informationon rational use of antibiotics comparable tothat available for the common companionand domesticated species. The situation issummarized well by Latt (1976) ' ...dosages of therapeutic agents for laboratoryanimals are scattered throughout thescientific literature or are extrapolated fromdosages recommended for other species'.

Other non-mammalian species are usedas laboratory animals in smaller quantities,for example birds and poikilotherms suchas reptiles, fish and amphibians and thesespecies will not be discussed as there can

Accepted 17 May 1994

be major differences in pharmacokinetics.An introduction to antibiotic therapeuticsfor these species can be found inspecialized medical texts (BSAVA, Jacobsenet al. 1991).

A number of areas will be considered:interference by antibiotics with anexperiment, antibiotic toxicity, routes ofadministration, effects of formulation onbioavailability, antibiotic prophylaxis, useof combinations of antibiotics, misuse ofantibiotics, regulatory approval forantibiotic use in animals, sources ofinformation on antibiotic indications anddose, and extrapolation of dose informationfrom other species.

Interference by antibiotics withexperimental studies

Bacterial infections in animals used forexperimental studies are undesirable. Atthe most basic level the morbidityproduced by an infection increases animaldiscomfort and experimental variation.More specific interactions can also occur,for example, studies on enteric pathogensor enteric diseases such as malabsorptionmay be compromised by concurrent entericbacterial infection and the associated

Laboratory Animals (1995) 29, 16-36


pathology. However, it should also berecognized that use of an antibiotic to'solve' the problem of a concurrentbacterial infection may itself interfere withthe experiment. The adverse effects of theaminoglycosides on renal function or thefluoroquinolones on juvenile cartilageformation may produce relatively obviousproblems. However, the effects may be lessovert. If a compound is underinvestigation, and concurrently anantibiotic with pharmacokinetic dispositionby the same mechanism is alsoadministered to the animal, then thekinetics of this compound underinvestigation may be affected. As a specificexample the concurrent administration ofchloramphenicol prolonged the duration ofxylazine-ketamine anaesthesia in rats butnot dogs (Nossmann et al. 1990).Concurrent administration ofchloramphenicol to dogs and catsundergoing pentobarbital anaesthesiasignificantly increased the duration ofanaesthesia (Adams & Dixit 1970). Theseresults suggest a competitive interaction atthe level of biotransformation.Chloramphenicol may competitivelyenhance the activity of bioinactivatedcompounds and decrease the activity ofbioactivated compounds metabolizedthrough the same family of cytochromesrequired for the metabolism of many drugs(Burns & Conney 1965). Direct druginteractions may also occur, fluoroquinolonescompete for gamma-amino butyric acid(CABA) receptor sites in the centralnervous system, and thus their use wouldbe contraindicated in certain neurologicalstudies (Bahri & Blouin 1991J. It is alsoclear from the examples cited above thatinterfering effects differ between species,making prediction more difficult. Studies ofantimicrobial efficacy may be compromisedwhere toxic antibiotics are used, such asclindamycin in the rabbit [Sande &Johnson 1975) without concurrent controlstreated with antibiotic but no bacterialchallenge. The confounding effects ofantibiotics can be even more subtle. Whenbacitracin, gentamicin and the antifungalnystatin were given to rats to reduce their

17

intestinal microflora, caecocolonic motilitywas altered and there was increased faecalexcretion of dry matter and water (Cherbutet 01. 1991). In a study that models a morecommon clinical situation, intestinalmotility was altered when amoxycillin-clavunate was given to human patients(Caron et 01. 1991J. These changes couldinfluence not only absorption and excretionof a test compound, depending on the sitesof drug absorption, but also studies of theintestinal tract itself. Antibiotic use shouldtherefore be very critically reviewed ifgiven during the course of an experimentalstudy.

Antibiotic toxicity

In all species even drugs with a high therapeuticindex can cause toxicity at very high dosesor when the usual dose is not adjusted totake account of the age and condition ofthe animal (Table 1). Studies with new-bornmice and guineapigs show that a varietyof liver enzyme systems are essentiallyabsent at birth and only begin to appear atthe end of the first week of life [Jondorfet 01. 1958). Renal excretion mechanismssuch as glomerular filtration and proximaltubular secretion may be absent at birthand develop over days or weeks dependingon species (Prescott & Baggott 1988). Inadult life impairment of renal or hepaticfunction, whether by disease or experimentalmanipulation, may alter the requiredantibiotic dose. As an example Table 1documents the adjustments on dosagerequired when using tetracycline inanimals with renal dysfunction. It mayeven be important to consider theanimals's diurnal rhythm (Heinze et al.1992). Mice are nocturnally active animals,and barbiturate sleep time can be up totwice as long when given during the dayrather than at night (Davis 1962).Biotransformation of some antibiotics maybe by hepatic mechanisms similar topentobarbitone. It is possible that doses for aparticular antibiotic might be extrapolatedfrom animals active during the day to themouse. When these agents are administeredto a mouse during the light phase they

18

Table 1 Modification of antibiotic dosage with reduced renal or hepatobiliary function

Morris

System affected

Hepatic function reduced

Biliary obstruction

Renal function reducedRenal function reduced

Renal function reduced

Renal function reduced

Antibiotic affected

Erythromycin, chloramphenicol, metronidazole,c1indamycin, lincomycin

Ampicillin, fluoroquinolones

Erythromycin, chloramphenicol, doxycyclineMost penicillins and cephalosporins,

c1indamycin, lincomycin, trimethorpim,sulphonamides

Aminoglycosides, carbenicillin, ticarcillin,vancomycin, metronidazole, fluoroquinolones

Tetracyclines, cephaloridine. nitrofurantoin,polymixin

Effect on dose

Reduce dose

Normally excreted in the bile,biliary obstruction may reduceaccessto site of infection

Give usual doseConsider minor dose reduction

Consider major dose reduction

Avoid using these antibiotics

Adapted from Prescott & Baggot 1988, Sande 1990b

could be metabolized to a different extentthan if given during the active dark phase,leading to differences in antibiotic plasmaconcentrations. This potential for differencesin plasma concentrations and even ineffectivetherapy would be increased if extrapolationof dose quantity or frequency was alsoinaccurate (see section below 'Extrapolationof antibiotic dose information betweenspecies'). In addition, antibacterials thatare eliminated via the kidney may haveblood levels that vary depending on thetime of day of administration. For examplesulphonamides given during the activephase in chickens and calves are eliminatedtwice as fast compared to when givenduring the animal's resting phase [Heinzeet al. 1992). There is also evidence ofstrain differences within laboratory specieswith respect to the toxic effects ofantibiotics. Tobramycin is more toxic inFischer rats than in Sprague-Dawley rats[Reinhard et al. 1991).

Many species have adverse reactions toparticular antibiotics at doses that arenormally safe therapeutic doses in otherspecies. In rodents and rabbits this is oneof the most important considerations inantibiotic therapeutics (Tables 2 and 3). Itshould be emphasized that the single mostimportant mechanism of antibiotic toxicityin rodents and rabbits is the secondaryeffects from the disruption of the normalenteric flora. In the guinea pig and hamsterthe specific cause of death is often the

toxin produced by overgrowing Clostridiwndifficile [Richard 1990, Fekety 1986,Manning et al. 1984). In one study inguineapigs, aureomycin caused overgrowthof Listeria monocytogenes which then ledto septicaemia and widely distributednecrotic lesions (Roine et al. 1953). In therabbit the toxins produced by C. perfringensand C. spiroforme have been implicated inlincomycin [Rehg & Pakes 1982) andclindamycin (Katz et al. 1978) inducedenteritis.

The hamster is particularly sensitive totoxigenic C. difficile overgrowth caused bya very wide range of antibiotics (Bartlettet al. 1978, Fekety 1986, Fekety et al.1979) and it has been suggested thattreatment of hamsters with any antibioticsshould be undertaken with caution sincemortality is high (Richard 1990).Tetracycline, metronidazole and to a lesserextent chloramphenicol are relatively poorinducers of enterocolitis in hamsters (seeTable 2) but the high incidence of renaldysfunction in older hamsters (Hubbard &Schmidt 1987) may even make tetracyclineadministration hazardous (see Table 1).

Apart from severe enterocolitis caused byother antibiotics even the 'classical' guineapigtoxicity to penicillin most probably is notdue to 'allergy' (Anon 1992). Newborn andgerm-free guineapigs are not susceptibleto penicillin toxicity (Manning et al.1984), so deaths in normal animals afterpenicillin may simply be another example


of antibiotic-induced enterocolitis. Toxicitymay be related to overgrowth of toxinsfrom Gram-negative bacteria (De Someret al. 1955), or C. difficile cytotoxin (Loweet al. 1980), which have been isolated fromcaecal contents of guineapigs which diedafter penicillin treatment. The incidence oftoxic effects may depend on whether suchorganisms are present in a particularanimal's flora. Thus guineapigs mayaccurately predict what happens in manwhen antibiotics are given at too high adose for too long. However, in anotherstudy (Roine et al. 1953) penicillinincluded in the diet at a rate of 50mg/kgdid not cause fatalities or weight loss.There do also remain reports of specificpenicillin hypersensitivity in guineapigs(Hoar 1976).

Many other antibiotics are also toxic inthe guineapig, [see Table 2), but certaincephalosporins at certain dose rates,cephaloridine 12.5mg intramuscularly for14 days (Dixon 1986), and cefazolin100mg/kg intramuscularly twice a day[Fritz et ai. 1987) have been shown not toproduce mortality.

Streptomycin is reported to be toxic tomice at doses of 3-6 mg/kg (Galloway1968, Harkness &. Wagner 1989a). Howevermassive single doses of streptomycin,500mg per animal, or 50mg per animalinitially followed by approximately 4mgdid not appear primarily to cause toxicity(Bonhoff et al. 19541.These higher dosesdid render animals more susceptible tosalmonella infection, but the mechanism isunclear. Anaerobic bacteria are recognizedas very important in limiting thecolonization of the digestive tract bypotential pathogens such as salmonella(Wiegersma et ai. 1982, Schaedler &. Orcutt1983) but streptomycin is inactive againstanaerobes.

Antibiotic toxicity in the gerbil hasreceived little attention, but in view of thesusceptibility to enterocolitis of some otherrodents, caution should be exercized. Incontrast, as summarized in a review byRichard (1990), the rat seems to tolerate awide range of antibiotics at therapeuticdose rates. As a specific example

19

lincomycin is toxic at relatively low dosesin guineapigs, hamsters (see Table 2) andrabbits (see Table 3), but 300mg/kg orallyto rats for 30 days was reported as non-toxic [Gray et ai. 1964). There has beenlittle stimulus for detailed studies on themechanisms of this relative tolerance inthe rat and mouse. Few clinical problemsare seen when many antibiotics are used inrat and mice, and the hamster is anexcellent model for antibiotic-inducedcolitis in man (Fekety 1986). One clue tothis tolerance may be found in studies byDabard et al. (19791,which suggest miceand rats are relatively resistant to theeffects of clostridia. After inoculation ofpregnant animals with Clostridiumdifficile, C. tertium and C. perfringens,large numbers of these bacteria were foundin the guts of their offspring. In rats andmice this caused no apparent problems,whilst in hares and rabbits fatal enteritisoccurred.

A wide range of antibiotics have beenreported as toxic in the rabbit (Table 3). Itis clear that lincomycin and clindamycinare particularly dangerous. Toxicity oferythromycin, spectinomycin and minocyclineis relatively milder, and is manifested asdepressed growth rates. The situation withpenicillins and cephalosporins is less clear.Ampicillin has been shown to cause seriousenteritis with mortality in several studies.The acute toxicity of penicillin is known,but where penicillin has been claimed tocontribute to enteritis other antibioticsthat cause enteritis, lincomycin (Thilsted1981) and ampicillin (Rehg &. Lu 19811,have also been fed. However thecombination of antibiotics can causemortality, for example aureomycin/sulphamethazine/penicillin, whereindividual agents did not (Hagen 19671.Indications that penicillin and cephalexinhave less potential to cause enteritis thanampicillin come from quantitative faecalbacteriological studies (Schroder et ai.1982). They reported the antibiotic-induceddepression in lactobacilli numbers, and therise in coliform bacteria and clostridialnumbers is less with penicillin andcephalexin than with ampicillin. There do

01~C,EooM1\\

ao;,.t::Iiit:oE

>-,t::Iiit:oE

uVI

01

Eoo'"oD-

OlEoo

N

'">-'""0

1Il

.S

..r::.•...~>-.t::Iiit:oE

•...IV•..E'":r

"" '">-'""0

00I

'"C

oD-

IVuCo

'"IM

C

>-,t::Iiit:oE

•...IVa.01",E '">-0'""'"0

-'"E·c'"

'"00>-'""0

>-'"..c "0>- IV

:t:::: -0

'" '"~ ,.,E ~~ ~oN III

V'I ~

~ r::r"0 ~IIIo IDIm.•...E"0 0'';;0UVI

01~C,EIIIN

~o~:J>Cou

..r::.•...~

"""" ~:Joo..r::.

<i'<:rONN C0'-D-'"IV::;)'"-00"000g ~~ •..

01a.'"IVC:J\:l

•..III '"•.. cr:I::QI'tl >-e Qj

,5•.."

'#. :J-u >-

III 0 CtI .~U 0'1 Iii~ .." ~ t:0 01- - 0:c ~>--.~ ~E';; ~ OI-l:: E ~ c,~ra :J

0 M 0 EO.•... ~ ci E 00 "'~IIIQIIII0'tl (5u '"'x c .~0 'g I:: C

I- .S I:: 'u I:: IV

a. I:: 'u >- ..r::.u c

VI :g I:: 'u >- E 'u c a.N '':; C IV .2 >- E 'E 'u E

.Q .S c E 0 >-QI ..c :g :~ '" IV .~ '" •.. ra E '"'" ..r::. 0 "0 a. •.. ..Q:ii '':; u a. a. -e '" u C IV C 0C 0 Era I:: IV IV '"

u C •... IV IV ..r::.I- <l: D- C: <l: u u i= :.::; U •.. \:l Z UVI

'~Nc. "Co ••~ .-••• "C.. '"Eiii'': E•...-OleE IV

oM NM;;;-

o"-01

-""Q;(5NIV)(

o~•..••E'"~.9-~01E•...

<0~

IVo

o"'o

...J

'$.oen

"'>.,'::IV1::oE

o"-+-01E

11'I

>.,'::iii1::oEo"-

NeQI,:::01•..oe

-iiiE'cIV

C,EoN

••eU>.u~•..

f!!.

<fiii'

01

-""C,Eo11'I

e'v>.Eo~:3

<J:

'¢

IM

,S~•..';:"C

••'ti'"iiiEe'"-o>.•...~';0-Eo"-01-""c.."'tE N

Vl

0>'o IV~"C

"'>.••• 0:.= ......•..•IV ••1:: .-o"CE 01

-""cteC1~o ::0. •••00 ••- 0 :.c

e

01

-""C,EMM

>..'::iii1::oE

e'v>.Ee~~w

~01

-""C,EooNIo

M

'">.,'::iii1::oE

22

Table 3 Adverse effects of antibiotic treatment in rabbits 1

Morris

Antibiotic

Ampicillin

Penicillin

Cephalexin

Lincomycin

Clindamycin

Tylosin

Erythromycin

Spectinomycin

Vancomycin

Minocycline

Spiramycin

Toxic dose

25 mg/kg 1M for 2 days5 mg/kg 1M for 2 days40 mg/kg for 4 days10 mg/kg PO for 6 days8 mg/kg bid 5C> 5 mg/kg PO antibiotic treated waterfor 3 days

lDso 5.25 g/kg PO

200 mg rabbitt for 7 days

100 mg PO single dose in 1.5-2.0 kgrabbits24 mg/kg PO antibiotic treated water30 mg/day PO in 2.0-2.5 kg rabbits1.3 mgl adult rabbit in feed for 3 days0.2 mg/kg 1M for 2 days

15 mg/kg PO for 3 days5 mg/kg PO for 2 days

Single IV dose of 30 mg/kg

100 mg/rabbit t for 7 days

3 gIl in drinking water for 7 dayst

1 gIl in drinking water for 7 dayst

75 mg/kg IV

30 mg/kg 1M for 3 days

acute oral LDso 4.85 g/kg

Toxic effects

Fata I enteritis 13Weight 105513

40% fatal enteritis over next 2 weeks'450% fatal enteritis over next month 1SEnteritis, previously also had penicillinsFatal enteritis in 7/11 rabbits 12

Both acute and chronic toxicity (enteritis)7

Diarrhoea 11

66% mortality with enteritis3


100% mortality with enteritis by 3 dayss30/130 rabbits died with enteritis9

33% mortality in 2 days'3


50% mortality with enteritis within 72hours 134/6 rabbits had fatal enteritis 12-14 daysafter treatment 16

Diarrhoea 11

Diarrhoea"

Diarrhoea"

Acute toxicity with 100% mortalitylO

Reduction in growth rate'S

Nervous signs 17

lData adapted from laval (1990). 2Boyd (1960), 3Rehg & Pakes (1982), 4Katz et al. (1978), sFesceet al. (1977), 6Maiers &Mason (1984), 7Boyd (1960), 8Rehg & lu (1981), 9fhilstead (1981), loNicolau et al. (1993), 11Schroderet al. (1982),12Milhaud et al. (1976), 13Ucois(1980). 14Morisse (1978), 1sSchatzmannet al. (1977). 16Upmanet al. (1992). 17Boyd& Price-Jones (1960). 18cited by laval (1990)PO=Per os1M= IntramuscularSC= SubcutaneousIV= Intravenoust=rabbits were 8-10 weeks old

not appear to be any detailed reports in theliterature on adverse effects of amoxycillin.Although its use in rabbits is usuallyproscribed in product information, this maybe at least partly based on potential fortoxicity. Manufacturers of amoxycillin dohear anecdotal reports of its use withoutapparent complications in rabbits (Hoare C,personal communication). However cautionwith amoxycillin is perhaps justified asshown by a recent single case investigatedby the author. A 4 kg rabbit was given7S mg of an injectable long actingamoxycillin preparation. Two days laterthe animal was found with circulatory

collapse and diarrhoea. At necropsy thecaecum was enlarged, contained fluidfaeces and its mucosa was haemorrhagic.Histological examination of the caecumshowed haemorrhage, mucosal ulcerationand heterophil invasion of the laminapropria and surface epithelium, similar tolesions reported in clindamycin-inducedenteritis [Katz et 01. 1978). Examination ofcaecal contents revealed a toxigenic strainof C. difficile to be present and ELISA(Launch Diagnostics) showed C. difficileenterotoxin was present. C. difficile toxinhas been reported in lincomycin-associatedcolitis in rabbits (Rehg & Pakes 1982). In


addition, all the aerobic bacteria isolatedwere coliforms, similar to the situationfollowing ampicillin administration(Schroder et ai. 1982).

As enterocolitis is such a problem itshould be noted that it has been preventedby oral administration of antibiotics thatare not absorbed across the intestinal tract.Gentamicin [80Itg/ml) and polymixin B(50ltg/ml) in the drinking water (Kaiseret al. 1992) have been reported as effectivein the guineapig. In hamsters bacitracin at3 mg/ml drinking water, or specificallywhen tetracycline (given at 500 mg/mldrinking water) is in use concurrentadministration of 250 mg/l of the non-absorbed sulphonamide sulphaguanidine,are reported as effective [Richard 1990).In the rabbit fatal enteritis normallycaused by ampicillin, (20mg/kg for 3 days)in the rabbit was avoided by concurrentadministration of gentamicin (10mg/kg/day)(Escoula et al. 1981), and enteritis causedby lincomycin (30mg/day orally for 3 daysin 2.0-2.5 kg rabbits) was prevented bygentamicin (30mg/ day orally) lFesce et al.1977). Another approach is to use ionexchange resins to bind the clostridialtoxins, and this has been shown to have abeneficial effect in the clindamycin-inducedenteroxaemia in rabbits lLipman et al.1992) and hamsters (Taylor &. Bartlett1980).

The fluoroquinolones are potentiallyuseful agents in veterinary medicine, asthey are broad spectrum, bactericidal,orally active and their is little cross-resistance developed to other classes ofantibiotics (Bahri &. Blouin 1991). Theseproperties alone make them of interest inlaboratory animal medicine, but therelative lack of incidence of inducedenterocolitis at clinical dose rates is note-worthy. Recently Rolf (1993) showed thatenrofloxacin, (22mg/kg) given to guineapigsorally for 6 days, produced no sign ofenterocolitis, in contrast to the enterocolitishe found only after a single injection ofpenicillin (60OOOiu).The reported doserange for enrofloxacin for rabbits, guineapigsand hamsters is in the range 5-10 mg/kgIDorrestein 1992). In rabbits 25mg/kg

23

enrofloxacin was given for 12 days as partof reproductive toxicity assessment withoutadverse effects (Althreuther 1992). The no-effect dose levels for rats and mice were165mg/kg and 550 mg/kg for 13 weeksadministration [Althreuther 1992).

Routes of administration

a) Oral administrationRelative to many other agents [see sectionabove on f Antibiotic toxicity'), tetracyclinesare often considered by clinicians to beuseful and relatively safe broad spectrumantibiotics for use in laboratory animals.Many authors (see section on drug dosesbelow) quote dose rates for oraladministration of tetracyclines. However,whilst these drugs may be safe in rats andmice (Dabard et al. 1979), hamsters(Bartlett et al. 1978, McNeil et al. 1986),gerbils (based on the author's experience ofcolony medication to control Bacilluspiliformis) and rabbits (Percy &. Black1988) both aureomycin (Roine &. Ettala1952), oxytetracycline (Roine et ai. 1953)and chlortetracycline (Eyssen et al. 1957)are toxic to guinepigs at therapeutic doserates. In addition in some of the specieswhere tetracyclines are safe, rabbits andrats, recent reports have shown that theoral route is of little use, and theireffectiveness in other rodents, wheresystemic absorption is required, should bequestioned. Administration of tetracyclinein the drinking water of rabbits atconcentrations up to 1600mg/l producedlow to undetectable serum levels, andwater intake was reduced at the highestdrug concentration (Percy & Black 1988].Even higher drinking water concentrationsof 4 gil tetracycline were given to rats andagain water intake was reduced and therewas no tetracycline detectable in theserum (Porter et ai. 1985). The therapeuticeffectiveness of oral tetracyclines inpreventing hepatic necrosis caused byTyzzer's disease in rodents (Harkness&. Wagner 1989b) should be viewed inthe knowledge that where presentB. piliformis is normally a resident of the

24

intestinal tract and that tetracyclines areprobably not absorbed and are thereforelikely to remain in the intestinal tract athigh concentrations.

Although amoxycillin is bactericidal, hasa broad antibacterial spectrum, is readilyavailable in formulations suitable foraddition to drinking water and as liquid fordirect oral administration, and is rapidlyand well absorbed orally (Mizen et a1.1981, Palmer et a1. 1976) toxicity in somerodent species may limit the use to ratsand mice and possibly rabbits (but seesection above on 'Antibiotic toxicity'). Thesame concerns apply to the less wellabsorbed penicillins.

Oral formulations of cephalosporins areavailable both as veterinary and humanliquids for direct oral administration. Theiruse has been suggested in several reviews(Flecknell 1983, McKellar 1989) however acritical review of the primary datasummarized in Tables 2 and 3, suggestedactual or potential toxicity of somecephalosporins in hamsters, guineapigs andrabbits, despite anecdotal reports of 'safe'clinical use. Cephalosporin use is morelikely to be safe in rats and mice, as manyreports of use in experimental infections,for example by Glauser and Bernard (1982),describe prolonged use of these drugs.

Oral presentations of trimethoprim/sulphonamides appear to be both relativelysafe and suitable for a broad range ofinfections (Richard 1990, Laval 1990). Mostformulations tend to be tablets, paediatrichuman products or veterinary productsformulated as pastes for administration tolarge animals. None of these presentationsis amenable to use as mass medication inthe drinking water, although water solublepoultry formulations could be utilized.

Chloramphenicol is available in oralpresentations and can be used inguineapigs. Given orally at doses between30 and 60mg/kg for 6 days it caused noovert toxicity (Eyssen et 01. 1957).However, to prevent drug resistance,chloramphenicol drug data sheets oftenrecommended that its use be restricted tosituations where clinical experience andlaboratory testing indicate no other

Morris

antibiotic can be used (National Office ofAnimal Health 1992). Alternatives inguineapigs include potentiatedsulphonamides and quinolones.

Fluoroquinolone antibiotics arepotentially useful as broad spectrum orallyactive antibiotics. Initial studies of oralabsorption of norfloxacin [dissolved insolvents normally only used inexperimental situations} in mice, rats andrhesus monkeys indicated effective bloodlevels could be achieved (Gilfillan et 01.1984).Pharmacokinetic studies on emofloxacinin rabbits have shown 5mg/kg twice aday orally gives effective tissue levels[Broome et 01. 1991). Dorrestein (1992) hassummarized the oral dose for rabbits,guineapigs and hamsters as 5-10 mg/kg,and the empirical drinking waterconcentration at 100mg/l.

b) Parenteral administrationOne of the most interesting aspects ofparenteral drug administration is what hasbeen termed 'the injection site dogma'(Marshall & Palmer 1980). This dogmaholds that peak serum concentrations aregreatest, and time of onset of this peakmost rapid when a drug is givenintravenously, and that intramuscularadministration gives greater peakconcentration and more rapid onset of peakserum concentration than subcutaneousadministration. Marshall and Palmer (1980)using calves and ten Voorde et 01. (1990)using dogs, injected ampicillin andamoxycillin subcutaneously andintramuscularly and found that this dogmawas not always true. There was oftenno major difference in bioavailability,especially with amoxycillin. Similarfindings were reported for long actingoxytetracycline in the rat (Curl et 01. 1988).Interestingly the site of subcutaneousinjection was often more important, thequantity of drugs absorbed over a fixedperiod of time was significantly less whenampicillin was given over the neckcompared to the thorax (ten Voorde et a1.1990). These findings have some impact insmall laboratory animals. Injectionvolumes may be relatively large, as some


rodent and rabbit doses are relativelyhigher than in the larger companionanimals (see section below on'Extrapolation of antibiotic doseinformation between species') for whichpreparations are formulated. The musclemass available for intramuscular injectionis small, so the risk of trauma, nervedamage, intermuscular injection and painon injection is greater. However, theexample given above may not hold for allantibiotics. The half-life of cefazolin in theguineapig is reduced by half when the drugis given by intramuscular compared tosubcutaneous injection [Kaiser et ai. 1992),which may reflect different absorptionkinetics. Intraperitoneal administrationmay also give higher drug levels thansubcutaneous administration. This wasshown in a study with oxytetracycline inrats, although therapeutic drugconcentrations were achieved via bothroutes and the irritant effect ofoxytetracycline was more marked via theintraperitoneal route (Porter et a1. 1985).Subcutaneous drug administration inrabbits and rodents technically is ofteneasier and safer and the questioning of theinjection site dogma does suggest thatbioavailability via this route may not bedissimilar to intramuscular injection.

Effect of drug formulation

Formulation, (salt form, solvents andexcipients) affects therapeutics of aparticular product by alteringbioavailability (the rate and extent ofabsorption of the same drug in the samedosage form). Bioinequivalence [defined asdifferences in bioavailability between theSame drug dosage form, for example 2brands of chloramphenicol tablets) betweenoral formulations of oxytetracycline,chloramphenicol (up to 100% differences inplasma concentrations) and penicillins«XI0 differences in peak concentrations)have been reported in dogs (Watson 1992).The use of long acting preparations ofoxytetracycline in the rabbit has beenrecommended by Laval (1990), they areroutinely used by the author in common

25

marmosets, and these preparations maymaintain effective oxytetracyclineconcentrations in the rat for at least 72h(Curl et ai. 1988). However, differentformulations of long acting oxytetracyclinediffer markedly in the degree of tissueirritation they produce. Those preparationsthat are most irritant can cause visibledistress to the animal, tissue damage,prolonged tissue residues and the lowestand most delayed peak concentrations(Nouws et al. 1990). Therefore care mustalways be exercised when transferringbetween different brands and in particularhuman generic presentations of antibiotics.A particular caution is necesssary withdifferent trimethoprim-potentiatedsulphonamide brands. They may containdifferent sulphonamides. For example,some veterinary brands contain sulfadiazineand in the rabbit the half life of thissulphonamide is only about 1h, comparedto 5-10 h in other species (DuSouich et a1.1978). Thus, depending on the brand, theeffective drug may be only trimethoprim.

Antibiotic prophylaxisThis is an important use and misuse ofantibiotics and Sande et ai.'s (1990a)classification of indications for prophylacticantibiotic use serve as a basis fordiscussion:-1. Prophylaxis to protect healthy animals

from infection. Examples in veterinarymedicine would be the well-recognizeduse of long acting oxytetracycline toprotect sheep exposed to pasteurellafrom infection (Frazer 1991). Inlaboratory animal medicine prophylaxismight include the use of oral tetracyclineto prevent outbreaks of Tyzzer's disease(Harkness &. Wagner 1989b). The linebetween use and misuse is often notclear. In some cases reduction ofexperimental stress might also preventinfection, but these changes may makeit impossible to use the animals for thatexperimental procedure.

2. Prevention of secondary bacterialinfection. Examples would includeprevention of bacterial infections in

26

immunocompromised animals orsecondary to viral disease (Walsh et 01.1988). Again whilst depopulation,decontamination and rederivation, orthe use of isolators may solve many ofthese problems, they may not befeasible or economical, especially in theshort term.

3. Prophylaxis prior to invasive proceduresin animals with previously implanteddevices. Antibiotics are used to preventcolonization of a previously implanteddevice (e.g. cathetersl, following thebacteraemia produced by minorprocedures (e.g. teeth cleaning orendoscopic intestinal biopsy). The sameconsiderations that apply to the timingof antibiotic administration at surgeryapply here.

4. Prophylaxis at surgery. The use ofantibiotics to prevent infection may be'the single most frequently abusedprinciple of (veterinary) surgery' (Rosin1988). If correct aseptic technique isfollowed there is usually no need forprophylactic antibiotics. However,perioperative wound contamination maybe inevitable. In many cases, even whensterile packs are prepared for rodentsurgery (McNeal et a1. 1988), surgicalprocedures are performed on multipleanimals using the same instruments. Inaddition, animal movement andinterferencewith the wound postoperativelyalso predisposes to infection. In realitymany rodent procedures are probablyperformed even without steriletechnique or antibiotic prophylaxis.This may be a result of the commonlyheld premise that rats are less susceptibleto perioperative infection than otheranimals. Recently this concept wascritically reviewed by Waynforth (1989).There was some evidence that rats mayeliminate bacteria more efficiently. Theprincipal factor in establishing woundinfection from wound contamination ishowever the concentration of bacteria(Goldschmidt 1972, Kaiser et a1. 1992).Waynforth (1989) cited the claim thatthe short procedures with smallincisional sites that are common in

Morris

rodent surgery might reduce theincidence or severity of infections thatwould otherwise be expected frominvestigators with limited surgical skillsand a reduced recognition of thepotential for infection. The argumentwould be that this low incidence maybe related to low bacterial numbers asthese incisions are smaller compared tothose in larger animals. Whilst it maybe true that the total number of bacteriaper wound might be less in rodentscompared to larger animals it could alsobe argued that the concentration ofpotential contaminants, principally theanimal's and operator's skin flora andunsterile equipment (Kaiser 1990),remains the same. Unfortunately thereare no well designed studies thatdirectly compare infection rates betweenspecies, using bacterial strains of thesame pathogenicity, similar inoculumvolumes, concentrations and sites, andwith similar co-factors such as foreignmaterial. These difficulties aredemonstrated by comparing methods indetail in studies by Goldschmidt (1972)using rats, by Kaiser et a1. (1992) usingguineapigs and by Elek &. Conen (1958)using human volunteers. It thereforeremains quite possible that the effects ofwound contamination in rodents arereal but also not easy to recognize. In arecent study craniotomy or laparotomywounds in rats were deliberatelyinfected during surgery (Bradfield et a1.1992). Despite the fact that the ratsshowed no clinical signs of infection,there were significant changes inmeasures of behaviour, serumbiochemical and haematologicalparameters and in the histologicalappearance of the wounds. Aseptictechnique should be used as standardbut there does seem to be a role for therational use of perioperative antibioticswhere aseptic technique cannot beabsolutely assured or inadvertant breaksin aseptic technique occurs.

The consensus regarding perioperativeantibiotic use in the human (Wenzel1992)andveterinary medical (Rosin 1988) literature


is that antibiotic use is recommended forclean-contaminated surgery and whenchronic implantation of foreign materialtakes place (Walsh et al. 1988). Whenwound contamination occurs duringsurgery it takes several hours for thebacteria to progress to a logarithmic phaseof growth, this is illustrated with datafrom an experimental study in the ratin Fig 1. It is therefore vital thatantibiotics, if they are used, should beadministered pre-operatively to allow peakblood levels to be achieved at the time ofwound contamination before significantbacterial proliferation takes place. Thechoice of antibiotic depends on the species,surgical procedure and local knowledge onantibiotic resistance, but it is likely thatthe chosen antibiotic needs to be given atleast 1-2 h preoperatively or intravenouslyat the onset of surgery (Wenzel 1992). Inaddition there is clear evidence that thereis no value in continuing antibiotics longerthan 24 h postoperatively (Weersink et al.1991, Haven et al. 1992), unless clinicalsigns suggest an infectious process resistantto the prophylactic regimen.

Antibiotics that have neuromuscularblocking properties (e.g. aminoglycosides,polypeptide antibiotics and possiblytetracyclines), should be used with greatcare when neuromuscular blocking agentsare used as part of the anaesthetic protocol(Jones 1992).

Use of antibiotic combinationsThe indications for use of combinations ofantibiotics include production of asynergistic effect, prevention of resistanceand efficacy in polymicrobial infections(Fantin & Carbon 1992.). However it isessential that these combinations arecritically reviewed (English & Prescott1983, Hirsch et al. 1990). In general thisreview has not occurred with rodents andrabbits and in addition the penicillins andaminoglycosides commonly used maycause toxicity [Tables 2 and 3). Someinformation on possible candidates forusable combinations can be gained fromstudies using rodents as animal models for

27

this application (Fantin & Carbon 1992,Mizen et al. 1991).

Misuse of antibioticsIt is worthwhile to highlight some of thecommon misuses of antibiotics (Sande etal. 1990a) to enable critical review ofantibiotic usage in rabbits and rodents.1. Treatment of untreatable infections.

This simply means using antibiotics insituations where they could never haveany effect on the infectious agent, forexample viral infections. This misuseshould be clearly distinguished fromacceptable use to treat a confirmedsecondary bacterial infection.

2. Therapy of non-specific pyrexia.Bacterial infections are not the onlycauses of pyrexia, they also includepain, postoperative pyrexia, metabolicdisorders etc.

3. Improper dosage. With smaller laboratoryanimals the main problem may be under-dosage, (see section below on 'Extrapolationof antibiotic dose information betweenspecies'), particularly if dosages areextrapolated on a mg/kg basis.

4. Absence of surgical drainage. When thereis purulent exudate there may simply beinadequate antibiotic penetration withoutsurgical drainage. An example herewould be pasteurella abscesses in rabbits.

5. Lack of adequate bacteriological information(e.g. sensitivity and blood levels). Inindividual cases this may mean relyingsolely on past experience and clinicaljudgement to select an antibiotic. Evenwhen species-specific bacteriology is notavailable or is pending it may help touse simple tests such as examination ofsmears for bacterial morphology andGram staining to select a more suitableantibiotic. The increasing use ofantibiotics in an unstructured mannercontributes to the development of drugresistant bacteria. Even in veterinarymedicine the development of strictpolicies to prevent nosocomial infections(defined as those acquired in animalholding or treatment facilities) and todefine antibiotic usage has been suggested

28 Morris

150,000

I3 ~ 100,000

.!1! J!l

.E; '§

.~ 0)

~ 'g.xl ;, 50,000

co'0.!1.

oo 24

- Sterile-0- In!ecled

48

TIme [hours)

Fig 1 Numbers of bacteria in rat peripheral blood aher introduction of a bacteria laden fibrin clot intothe abdominal cavity via a laparotomy (De Marsh P & Smith E, personal communication)

Data from: 'Jones (1986), 2Wiegersma et al. (1982).

Table 4 Effects of antibiotics on host microfloraand colonization resistance

(Murtaugh &. Mason 1989).An indicationof the potential of different antibioticsto produce resistant bacteria bydisturbing the resident flora is given inTable 4, although there may be speciesdifferences (Wiegersma et a1. 1982).

Regulatory approval for use ofantibiotics in laboratory animalsWhen treating the common domesticanimals many of the factors mentionedabove are taken into account and combinedwith the dose obtained from the product'slabelling. This option is usually notavailable when treating laboratory species.It has been suggested that the pharrnacokineticdata obtained from rodents or rabbits in theearly stages of a <!.rug'sdevelopment be

submitted to the regulatory authorities andthis would provide dose information atlittle cost. There are at least two problemswith this approach. First, many of thesestudies may not be at therapeutic doserates and the formulations of the drug areoften not comparable (Gilfillan et a1. 1984).Second, the regulatory authorities requireconsiderably more data than just thisinformation. In a large market such asthe United States, for example (Kilgore R,personal communication, Guest 1993) thepharmacokinetic data merely provides theinitial data for a dose titration infectionstudy in the target species, and this mustbe followed by a second well-controlledinfection study. If required, for example ifindications are sought for administration torabbits reared for food, the most costlyparts of the process are the safety studies,consisting of chronic toxicity testing at 3dose rates and an acute toxicity test, thesecosts alone can exceed ECU 100000.Nevertheless in certain markets it ispossible that the increasing numbers ofsmall mammals being kept as pets mayencourage the development of productswith specific labelling suitable forlaboratory species. In France for example arecent survey showed 7.6 million households(6% of the total) possessed rodents, comparedto 33% with a dogand 22% with a cat, and thishas encouraged pharmaceutical companiesto investigate this market (Anon 1993).

Cepha lospori ns1.2, am inog Iycosides I,

trimethoprim '.2, sulphonamides1.2,erythromycin '.2, doxycycline 1.2,

parental penicillins'

Amoxycillin 1.2, tetracycline',chloramphen icol'

Ampicillin', cloxacillin 1.2,

metronidazole', furazolidone',penicillin V2

No adverseeffects

Moderateeffects

Significanteffects


Sources of information in drugaddiction and dose

Apart from antibiotic doses listed astreatment of specific diseases in clinicaltexts, there are a number of reviews thatcan act as a guide for clinicians. Aselection of some of the most usefulis given below. A comprehensive referencedlist of doses for rats is given byKruckenberg (1979), this list includesantibiotics, although only one-third ofthe citations refer to primary studieson the antibiotic. Indications, appropriateantibiotics and their doses for rodentsare reviewed by Richard (1990), for rabbitsby Laval (1990) and Mureau (1988) andfor rodents and rabbits by Jacobsen et ai.(1991). Doses for rabbits, rodents andother species are also given by Latt(1976), Flecknell (1983), Harkness& Wagner (1989a) and McKellar (1989).

A Formulary for Laboratory Animals byC. Terrance Hawk and Steven L. Learywhich will contain fully referenced doses isplanned to be published by Iowa StateUniversity Press, Ames, Iowa in May 1995.

Extrapolation of antibiotic doseinformation between species

In the majority of situations an antibioticdose cannot be found for a particularlaboratory species, or the basis on whichthe published dose was obtained is unclear.In particular there is often no convenientway to discover whether the data is obtainedor extrapolated from primary studies on thesame species or if it was extrapolated orestimated from doses in other species. Anunderstanding of the basic principles forthe extrapolation of doses between specieswould therefore be useful.

There are two general approaches to inter-species scaling, either using an allometricor a physiological approach (Mordenti 1986,Mizen & Woodnutt 1988).

Allometry is the study of the proportionchanges correlated in variations in size ofeither the total organism or the part underconsideration (Boxenbaum 1984). It hasbeen found that over 100 highly diversebiological parameters and many

29

pharmacokinetic parameters are linearlyrelated to body size (Calabrese 1991),3 indicators have been found useful ininterspecies scaling: body weight, bodysurface area and physiological time(Table 5).

For body weight the mathematicalprinciple that underlies this allometricrelationship is described by the equation:

Log P = Log a + b.Log W, which canbe simplified tOj P = aWb

Where P = parameter of interest, W = weight,with a and b as constants. This formula isdescribed and illustrated by Kirkwood(1983). The constant b has been found formany parameters to be approximately 0.75(Kirkwood 1983, Calabrese 1991). Theconstant a fixes the value of P whenW = 1kg (Kirkwood 1983, Mordenti 1986).A special case may exist for very smallmammals with a weight between 2.5 and100 g, where the constant b may be lowerat 0.5 (Bartels 1982). These models can beused to clarify and predict relationshipsbetween dose and body weight betweendifferent species (Kirkwood & Merriam1990). How in practice can results such asthose in Table 6, which are calculated fromactual data, be derived when doseinformation for a particular species is notavailable? An example is shown in Table 7to clarify the process. It is immediatelyapparent that in many cases the doseextrapolated on a mg/kg°.75 basis,especially for smaller animals such asrodents, is relatively higher than if it wereextrapolated from the dose for a largeranimal such as a companion animal on amg/kg basis. The advantages of thesubcutaneous route of injection with thisrelatively higher dose volume is thereforeeven more obvious.

Body surface area has been used insteadof body weight to derive allometriccomparisons. The use of this indicator isperhaps best known in medicine forantineoplastic drugs, and an example using3 species of laboratory rodents and man isshown in Table 8. Identical sulphadiazineblood levels were achieved in humanpatients when doses were calculated on abody surface area basis (Calabrese 1991).

30 Morris

Table 5 Representative body surface area to body weight rate ratios forvarious species

Body weight Surface area Dose equivalentSpecies (kg) (m2) Km factor" (kg-1)

Man, adult 60 1.6 37.5 1Man, child 20 0.8 25 1.5Mouse 0.02 0.0006 3 12.5Rat 0.15 0.025 6 6.3Cat 3 0.24 12.5 3Dog 16 0.65 24.5 1.5Sheep/goat 50 1.1 45.5 0.8Pig 75 1.5 50 0.75Cow 150 2.4 62.5 0.6Cow 500 5.0 100 0.4Pony 280 4.4 63.5 0.6Horse 350 4.0 87.5 0.4Horse 650 5.9 110 0.3.

aTo express a mg/kg dose in any given species as an equivalent mg/m2 dose, multiply thedose by the appropriate Km factor. E. g. in the cat 10 mg/kg is equivalent to10 mg/kgx 12.5= 125 mg/m2•Setting the dose equivalent for man (adult, 60 kg) as 1, we may obtain the doseequivalent kg-I (in relation to man) dividing the Km factor for man by the Km factor ofany species given in the table.Data from Van Miert (1989). reproduced with permission

1.Weight Suggested dose

Species (kg) mg/kg mg/kgo.75

Horses and 500 6.94 32.50cattle 2.

Calves, pigs 100 11.36 35.00and sheep

Piglets 10 25.00 45.00Dogs and cats 3 75.00 32.50

Table 6 Comparison of streptomycin/dihydro-streptomycin 1 dose rates between species on amg/kg and mg/kgo.7s basis

1a mixture of 125 mg/ml of each drug, total 250 mg/mlactive agents.Data adapted from Kirkwood (1983)

Body surface area, calculated using theallometric equation:-

Surface area = 11.7 x Weighto.66

was used to extrapolate doses of antibioticsfrom man to rhesus monkeys (Kelly et al.1992). However, the surface area methodhas not been used to any great extentbecause of controversy over whether it isfundamentally more inaccurate than bodyweight (Calabrese 1991). An example is thedifficulty in measuring surface area in an

Table 7 Worked example to extrapolate oral doseof amoxycillin from dog to squirrel monkeys

Dose in dog = 10 mg/kg (National Office ofAnimal Health 1992)Total dose for 10 kg dog=100 mgDose for a 10 kg dog expressed inmg/kgo.75 = 17.8 mg/kgo.7s

For a 750 9 squirrel monkeyTotal dose =weighto.75 x dose inmg/kgO.75=0.80 x 17.8 mg/kgo.7S= 14.34 mg(Total dose if extrapolation had been on a mg/kgbasis would be 7.5 mg)Converting this total dose of 14.34 mg for a750 9 squirrel monkey to mg/kg:-Dose mg/kg=19.12 mg/kg

3. Comparison to actual pharmacokinetic data(Mizen et al. 1981)Peak blood level for dog given oral dose of10 mg/kg amoxycillin =6.1I'g/mlArea under the curve for dog given oral dose of10 mg/kg amoxycillin = 15.3I'g.h/mlPeak blood level for squirrel monkey given oraldose of 25 mg/kg amoxycillin = 13.0 I'g/mlArea under the curve for squirrel monkey givenoral dose of 25 mg/kg amoxycillin = 29.5 I'g.h/ml

individual (Van Miert 1989). However, VanMiert (1989) has correlated the product ofbody surface area and body weight andproduced a more accurate basis for


Table 8 Comparison of the mg/kg and mg/m2 dose of the antineoplastic drugmechlorethamine in man and 3 laboratory rodents

Surface Total dose Total dose Total dose Total doseSpecies Weight (kg) area (m2) (mg) (mg/kg) (mg/m2) (mg/m2) *

Mouse 0.018 0.0075 0.072 4.0 9.6 12Hamster 0.050 0.0137 0.15 3.0 10.9Rat 0.25 0.045 0.5 2.0 11.1 12Man 70.0 1.85 21-28.0 0.3-0.4 11.3-15.1 11.2

Data from Pinkel (1958), except * from Van Miert (1989), see Table 5, this dose was obtained by using theKm factor multiplied by the dose in mg/kg

31

interspecies comparison, see Table 5. Manychronological parameters, such as life span,number of heartbeats per minute andgestation period also are related by thebasic allometric equation described above,but in this case the constant b averagesapproximately 0.25 (Kirkwood 1983). Thismeans that as body size decreases theseparameters greatly increase and this is wellillustrated in Fig 2. The implication of thisis that instead of raising the dose rate inproportion to the dose, the dose frequencycan be increased with timeO.25. However invery small rodents this may be impracticalas the dose frequency becomes too high(Fig 2).

The allometric approaches describedabove are inherently simple andapproximate, indeed they have beendescribed as a 'black box' approach,because no attempt is made to determineorgan distribution or make physiologicalassumptions (Mordenti 1985 & 1986).In contrast the physiological approach ismore complex. Blood flow to eliminatingorgans, tissue and fluid volumes, drugconcentrations, protein and enzymebinding are measured in one species.Conceptual models are drawn to describethe presence and movement of drugs andmetabolites within body compartments.Mass balance equations are then drawn forthe sum of the processes occurring in eachcompartment and solved simultaneously.Once pharmacokinetics are definedpredictions for other species are obtainedby substituting the original biochemical,anatomical and physiological data withdata from the new species and recalculatingthe equations to produce extrapolated

pharmacokinetic data [Mordenti 1985 &1986, Baggot 1992).

It is obvious that basic allometry is themost simple method of species scaling. Itis possible to develop more complexallometric models to derive more detailedpharmacokinetic data (Boxenbaum 1984)and overcome the limitations of the simplemg/kgO.75 approach. However, certainconditions have been suggested forsuccessful use of allometry.1. Compounds are renally excreted2. There are no interspecies differences in

metabolism3. Protein binding is low4. Pharmacokinetics are first order5. Confirmatory data is obtained from a

wide range of species and body weights(Mizen & Woodnutt 1988)

Unfortunately these conditions are notalways all met, and in particular speciesdifferences in metabolism are notuncommon, such as trimethoprim (Baggot1992), sulphadiazine (De Souich et al. 1978)or novel antibiotics (Smith et al. 1973).

The physiological approach is the methodof choice where the details of drugdistribution are important and where thereis strong protein binding and extensivemetabolism. However, the methodology ismore complex and costly (Ritschel et al.1992), less commonly used and usuallycalculated using computer programs.

ConclusionsAntibiotics may interfere with anexperimental protocol, either by a directinteraction or by influencing metabolism orpharmacokinetics of compounds under

32

90

CHRONOLOGICAL CLOCK

CefitoximeHalf-Life(minutes)

30 7500

BIOLOGICAL CLOCK

CefitoximeHalf-Life

(heartbeats)

Morris

2500

5000Fig 2 Ceftizoxime half-life in various mammals depends on the reference system used to denote time.(a) Chronological clock for ceftizoxime half-life based on chronological time. Half-lifes are reported inminutes (b) Biological clock for ceftizoxime half-life based on physiological time. Half-Iifes are reported inheartbeats. From Mordenti (1985) with permission. ©1985 American Society for Microbiology

investigation. The incidence of this isprobably underestimated and antibiotic usemust be critically reviewed. It is clear thatthe principal side effect of antibiotics inrodents and rabbits is induced enterocolitis.The resistance of rats and mice to thisphenomenon is also remarkable. Moreconsideration should be given to the routeof antibiotic administration to rodents andrabbits. Suitable agents and presentationsdo exist for effective oral administration,but these do not include commontetracyclines, and as an alternativefluoroquinolones show promise as effectiveand safe broad spectrum agents. Bearing inmind the small muscle mass of rodents andrabbits and the relatively large injectionvolume, compared to larger species,coupled with evidence that thesubcutaneous route may be as effective asintramuscular injection the former routeshould always be considered. The indicationsfor antibiotics should also be criticallyreviewed, as in other areas they are probablyoverused. In particular prophylaxis atsurgery may be less than optimal as rationalguidelines are still not being followed. Aconsideration of the basic principles ofveterinary pharmacology and the factors

summarized above will help rationalize theuse of antibiotics in laboratory species.

The lack of controlled studies onantibiotic therapeutics in laboratoryanimals, together with relatively littleextrapolation from primary data that isavailable, means that many 'doses' arebased on clinical response and lack of overtside effects. It is possible that as smallmammals become more popular as petsspecific products, with dosage informationbased on controlled studies, may becomeavailable. However a complete range ofsuitably labelled products for all specieswill never be available. Where justified theuse of scaling with doses extrapolated fromother species, particularly as it is practicalusing allometric principles, should result inmore realistic dose regimens.

Acknowledgments I would like to thank A. Gisby,R. Kilgore, 1. Mizen, G. Palmer &. 1. Rolf for theirassistance and P. Flecknel1, H. Hughes, M. Landi,1. Rolf and B. Waynforth for critical review.

ReferencesAdams RA, Dixit BN 119701 Prolongation of

pentobarbitol anaesthesia by chloramphenicol indogs and cats. Toumal of the American VeterinaryMedical Association 156, 902-6


Altreuther P (1992) Safety and tolerance ofenrofloxacin in dogs and cats. In: Proceedings ofthe 1st Intemations Baytril Symposium Bonn:Bayer AG, pp 15-19

Anon (19921 Animal rights lobby stung by theexposure of penicillin myth. Research DefenceSociety Newsletter April 11

Anon (1993) New elass of pet opens marketopportunities in France. Animal Pharm no. 284,5-6

Baggot JD (1992) Clinical pharmacokinetics invcterinary medicine. Clinical Pharmacokinetics22, 254-73

Bahri LE, Blouin A 119911Fluoroquinolones: A newfamily of antimicrobials. The Compendium ofContinuing Education for the PracticingVeterinarian 13, 1429-33

Bartels H (1982) Metabolic rate of mammals equalsthc 0.75 power of their body weight.Experimental Biological Medicine 7, 1-11

Bartlett JG, Chang TW, Moon N, Onderdonk AB(1978) Antibiotic induced lethal enterocolitis inhamstcrs: Studies with eleven agents andevidence to support the pathogenic role of toxinproducing bacteria. American Journal ofVeterinary Research 39, 1525-30

Bohnhoff M, Drake BL, Miller CL 119541Effect ofstreptomycin on susceptibility of intestinal tractto experimental salmonella infection. Proceedingsof the Society for Experimental Biology andMedicine 86, 132-7

Boxenbaum H (1984) Interspecies pharrnacokineticscaling and the evolutionary-comparativeparadigm. Drug Metabolism Reviews 15,1071-121

Boyd CE (1960) The acute oral toxicity ofbenzylpenicillin potassium in the rabbit.Antibiotics and Chemotherapy 10, 376-84

Boyd EM, Price-Jones MA {19601 The comparativeacute oral toxicity of spiramycin adipate in mice,rats, guineapigs and rabbits. Antibiotics andChemotherapy 10, 273-84

Bradfield JF, Schacht man TR, McLaughlin RM,Stcffcn EK (19921 Behavioural and physiologicaleffects of inapparent wound infection in rats.Laboratory Animal Science 42, 572-8

Broome RL, Brooks DL, Babish JG, Copeland DD,Conzelman GM (1991) Pharmacokineticproperties of cnrofloxacin in rabbits. AmericanJournal of Veterinary Research 52, 1835-41

Burns IT, Conney AH (19651 Enzyme stimulationand inhibition in the metabolism of drugs.Proceedings of the Royal Society of Medicine 58,955-60

BSAVA (Publication dates varyl Manuals of;Reptiles, Ornamental Fish, Parrots, Budgerigarsand other Psittacine Birds, Exotic Pets.Cheltcnham UK: British Small Animal VeterinaryAssociation

33

Calabrese EJ (1991) Scaling: An attempt to find acommon denominator. In: Principles of AnimalExtrapolation Chelsea, Michigan: Lewis,pp 499-527

Caron I, Ducrotte P, Lerebours E, Colin R,Humberts G, Denis P (1991) Effect of amoxycillin-clavulanate combination on the motility of thesmall intestine in human beings. AntimicrobialAgents and Chemotherapy 35, 1085-8

Cherbut C, Ferre JP, Corpet DE, Ruckcsbusch Y,Delfort-Laval J (1991) Alterations of intestinalmicroflora by antibiotics: Effects on fccalexcretion, transit time and colonic motilityin rats. Digestive Diseases and Science 36,1729-34

Curl JL, Curl JS, Harrison JK (1988) Pharmacokineticsof long acting oxytetracycline in the laboratoryrat. Laboratory Animal Science 38, 430-4

Dabard 1, DuBois R, Martinet L, Ducluzcan R(1979) Experimental reproduction of neonataldiarrhoea in young gnotobiotic haressimultaneously associated with Clostridiumdifficile and other clostridium strains. lnfectionand Immunity 24, 7-11

Davis WM (1962) Day-night periodicity in pentobarbitalresponse of mice and the influence of socio-psychological conditions. ExperimentiaXVID/515,235-7

De Somer P, Van de Voorde H, Eyssen H, VanDijck P (1955) A study on penicillin toxicity inguinea pigs. Antibiotics and Chemotherapy 5,463-9

Dixon LW 119861 Antibiotic toxicosis in the guineapig. Texas Veterinary Medical Journal 48, 31

Dorrestein GM 119921 Enrofloxacin in pet avian andexotic animal therapy. In: Proceedings of the 1stInternational Baytril Symposium Bonn: BayerAG, pp 63-70

DuSouich P, McLean Al, Lalka D, Jenkins B,Haegele DK, McNay JL (1978) Sulfadiazinehandling in the rabbit. 1. Pseudosaturation of N-Acetyltransferase. Journal of Pharmacology andExperimental Therapeutics 207, 221-35

Ebright JR, Fekety R, Silva 1, Wilson KH 119811Evaluation of eight cephalosporins in hamstercolitis model. Antimicrobial Agents andChemotherapy 19, 980-6

Elek AD, Conen PE (1958) The virulence ofStaphylococcus pyogenes for man. A study of theproblems of wound infection. British Journal ofExperimental Pathology 38, 573-86

English PB, Prescott CW (1983) Antimicrobialchemotherapy in the dog. II: Some practicalconsiderations. Joumal of Small Animal Practice24, 371-83

Escoula L, Canguilhem R, Larrieu G, More J (19811Sur la sensibilite du lapin a I' associationanpicillin-gentamicine. Annales RecherchesVeterinaire 12, 11-7

34

Eyssen H, De Somer P, van Dijck (1957) Furtherstudies on antibiotic toxicity in guinea pigs.Antibiotics and Chemotherapy 7, 55-61

Fantin B, Carbon C (1992) In vivo antibioticsynergism: Contribution of animal models.Antimicrobial Agents and Chemotherapy 36, 907-12

Farrar WE, Kent T H, Elliott VB (1966) Lethal gram-negative bacterial superinfection in guinea pigs givenbacitracin. Journal of Bacteriology 92, 496-501

Fekety R, Silva 1, Toshniwal Ret al. (1979)Antibiotic associated colitis: Effects of antibiotics onClostridium difficile and the disease in hamsters.Review of Infectious Diseases 1, 386-96

Fekety R (1986) Animal models of antibiotic-induced colitis. In: Experimental Models inAntimicrobial Chemotherapy Volume 2 NewYork: Academic Press, pp 61-72

Fesce A, Ceccarelli A, Fresce E, Balsari A (1977)Ecophylaxis: Preventive treatment withgentamicin of rabbit lincomycin-associateddiarrhoea. Folia Veterinari Latino 7, 225-42

Fleckncll P (1983) Retraint, anaesthesia and treatmentof children's pets. In Practice May, 85-95

Fraser CM [cd) (1991) Septacemic pasteurellosis ofsheep. In: The Merck Veterinary Manual Rahway NJ:Merck &. Co, p 401

Fritz PE, Hurst WI, White WJ, Lang CM (1987)Pharmacokinetics of cefazolin in guineapigs.Laboratory Animal Science 37, 646-51

Galloway JH (1968) Antibiotic toxicity in white mice.Laboratory Animal Care 18, 421-5

Gilfillan Ee, Pelak BA, Bland JA, Malatesta PF,Gadebusch HH (19841 Pharmacokinetic studies ofnor£loxacin in laboratory animals. Chemotherapy30, 288-96

Glauser MP, Bonard M (1982) Treatment of ascendingEscherichia coli pyelonephritis with Ceftriaxonealone and in combination with gentamicin.Chemotherapy 28, 410-6

Goldschmit F (1972) Reproducible topical staphylo-coccal infection in rats. Applied Microbiology 23,130-4

Gray JE, Purmalis A, Feenstra ES (1964) Animaltoxicity studies of a new antibiotic lincomycin.Toxicology and Applied Pharmacology 6, 476-96

Guest GB (1993/ Using proper drugs and usingthem poorly. Agri-Practice 14, 20-4

Hagen KW (1967) Effect of antibiotic-sulfonamidetherapy on certain microorganisms in the nasalturbinates of domestic rabbits. Laboratory AnimalCare 17, 77-80

Harkness JE, Wagner JE (1989a) ClinicalProcedures. In: The Biology and Medicine ofRabbits and Rodents 3rd edition. Philadelphia:Lea &. Febiger, pp 55-84

Harkness JE, Wagner JE (1989b) Tyzzer's disease.In: The Biology and Medicine of Rabbits andRodents 3rd edition. Philadelphia: Lea &. Febiger,pp 198-200

Morris

Haven ML, Wichtel JJ, Bristol DG, Fetrow JF,Spears JW (1992) Effects of antibiotic prophylaxison post operative complications afterrumenotomy in cattle. Journal of the AmericanVeterinary Medical Association 200, 1332-5

Heinze W, Kruger S, Schroder A (19921 Chronologicalinfluences on the action of drugs. Monatsheffe {iirVeterinarmedizin 47, 643-51

Hirsch DC, Jang SS, Biberstein EL (1990) Lack ofsupportive susceptibility data for use of ampicillintogether with trimothoprim-sulfonamide as abroad-spectrum antimicrobial treatment ofbacterial disease in dogs. Journal of the AmericanVeterinary Medical Association 197, 594-6

Hoar R (1976) Toxicology and Teratology. In: TheBiology of the Guinea-pig [Wagner 1, Manning P,edsj. New York: Academic Press, p 273

Hubbard G B, Schmidt RE (19871 Noninfectiousdiseases. In: Laboratory Hamsters (Van Hoosier G,McPherson C, cds). New York: Academic Press,pp 169-72

Jacobsen E, Kollias GV, Peters 11 (19911 Dosages forantibiotics and parasitacides used in exoticanimals. In: The Compendium Collection ofContinuing Education for the PracticingVeterinarian (Johnston DE, edl. Trenton NJ:Veterinary Learning Systems, pp 202-9

Jondorf WR, Maickel RP, Brodie BB (1958) Inabilityof newborn mice and guinea-pigs to metabolisedrugs. Biochemical Phannacology 1, 352-4

Jones RL (1986) Control of nosocomial infection. In:Current Veterinary Therapy IX (Kirk RW, edl.Philadephia: WB Saunders, pp 19-24

Jones RS (1992) Muscle relaxants in canineanaesthesia 2: Clinical applications. Journal ofSmall Animal Practice 33, 423-9

Kaiser AB (1990) Post operative infections andantimicrobial prophylaxis. In: Principles andPractices of Infectious Disease 3rd edition(Mandell EL, Douglas RG, Bennet JE, cds). NewYork: Churchill Livingstone, pp 2245-57

Kaiser AB, Kernodle DS, Parker RA 11992) Low-Inoculum model of surgical wound infection.Journal of Infectious Diseases 166, 393-9

KaipainenWJ, Faine S (1954 Toxicity oferythromycin. Nature 174, 969-70

Katz L, LaMont JT, Trier JS, Sannenblick EB,Rothman SW, Broitman SA, Rieth S (19781Experimental clindamycin associated colitis inrabbits. Gastroenterology 74, 246-52

Kelly D1, Chulay 10, Mikesell P, Friedlander AM (1992)Serum concentrations of penicillin, doxycycline andcipro£loxacin during prolonged therapy in rhesusmonkeys. Journal of Infectious Diseases 166, 1184-7

Killby VA, Silverman PH (1967) Toxicity ofantibiotics in laboratory rodents. Science 156, 264

Kirkwood JK (1983) Influence of body size onanimals on health and disease. Veterinary Record113, 287-90


Kirkwood JK, Merriam J [1990) Variation in plasmahalf life of gentamicin between species in relationto bodyweight and taxonomy. Research inVeterinary Science 49, 160-5

Knoop FC (1979) Clindamycin associatedenterocolitis in guinea pigs: Evidence for abacterial toxin. Infection and Immunity 23,31-3

Kruckenberg SM (1979) Drugs and Dosages. In:The Laboratory Rat Vol. 1 [Baker H, Lindsey J,Weisbroth SH, eds). New York: Academic Press,pp413-21

Latt RH (1976) Drug dosages for LaboratoryAnimals. In: Handbook of Laboratory AnimalScience Vol III (Melby EC, Altman NH, eds). VolIII, Cleveland, Ohio: CRC Press, pp 561-9

Laval A (1990) Choix de l'anti-in£ectieus chez Ie lapind'agremeDt.Recueil de Medicine Veterinaire 166,375-9

Licois 0 (1980)Action toxique de certains antibiotiqueschez Ie lapin. Recueil de Medicine Veterinaire 156,915-9

Lipman NS, Weischedel AK, Connors MJ, Olson DA,Taylor NS (1992) Utilisation of cholesrryramineresin as a preventative treatment for antibiotic(clindamycin) induced enterotoxaemia in therabbit. Laboratory Animals 26, 1-8

Lowe BR, Fox JC, Bartlett JC (1980) Clostridiumdifficile associated cecitis in guinea pigs exposedto penicillin. American Toumal of VeterinaryResearch 41, 1277-9

Maiers ro, Mason SJ (1984) Lincomycin-associatedenterocolitis in rabbits. Toumal of the AmericanVeterinary Medical Association 185, 670-1

Manning PJ, Wagoer JF, Harkness J E (1984) Biologyand diseases of guinea pigs. In: Laboratory AnimalMedicine (Fox J, Cohen B, Loew F, eds). NewYork: Academic Press, p 173

Marshall AB, Palmer CH (1980) Injection sites anddrug bioavailability. In: Trends in VeterinaryPhannacology and Toxicology, Proceedings of the 1stEuropean Congress on Veterinary Phannacology andToxicology [van Miert A, Frens J, van der Kreek F,cds). Amsterdam: Elsevier, pp 54-60

McKellar QA (19891Drug dosages for smallmammals. In Practice March, 57-61

McNeal J, Brown MJ, Bennet BY (1988) Sterile drapepacks for survival surgery in small animals. LaboratoryAnimal Science 38, 517-18 Abstract p 33

McNeil PE, AI-Mashat FR, Bradley RA, Payne AP(1986) Control of an outbreak of wet tail in aclosed colony of hamsters. (Mesocricetus auratus).Veterinary Record 199, 272-3

Milhaud C, Renault L, Vaissarie 1, Maire CI [1976)Scnsibilite du lapin a l'ampicillin. Recueil deMedicine Vt3terinaire 152, 843-7

Mizen L, Bhandari K, Sayer J, Catherall E [1981)Pharmacokinetics and distribution of Augmentinlamoxycillin/clavulanic acid) in laboratory animals.Drugs under Experimental Clinical Research VO,263-7

35

Mizen L, Woodnutt C (1988) A critique ofanimal pharmacokinetics. Toumal ofAntimicrobial Chemotherapy 21, 273-80

Mordenti J (1985) Forecasting cephalosporin andmonobactam antibiotic half-lives in humans fromdata collected in laboratory animals.Antimicrobial Agents eJ Chemotherapy 27,887-91

Mordenti J (1986) Man versus beast: Pharrnacokineticscaling in mammals. TournaI of PharmaceuticalSciences 75, 1028-40

Morisse J-p (1978) Induction d'une enterite de typecolibacillaire chez Ie lapin. Recueil de MedicineVeterinaire 129, 625-32

Mureau (1988) L'antibiothCrapie chez Ie lapin.Cuniculture IS, 75-9

Murtaugh RJ,Mason CD (1989)Antibiotic pressurc andnosocomial diseasc. Veterinary Clinics of NorthAmerica: Small Animal Practice 19, 1259-74

National Office of Animal Health 119921Compendium of Data Sheets for VeterinaryProducts. London: Datapharm Publications

Nicolau DP, Freeman CD, Nightingae CH,Quintiliani R (1993) Pharmacokinetics ofminocycline and vancomycin in rabbits.Laboratory Animal Science 43 222-5

Nossman BC, Amouzadeh HR, Sangiah S (1990)Effects of chloramphenicol, cimetidine andphenobarbital on and tolerance to xylazine-ketamine anaesthesia in dogs. Veterinary andHuman Toxicology 32, 216-9

Nouws JFM, Smulders A, Rappalini M (1990)A comparative study on irritation and residueaspects of five oxytetracycline formulationsadministcrcd intramuscularly to calves, pigsand sheep. The Veterinary Quarterly 12,129-38

Palmer CH, Buswell JF, Dowrik JS, Yeoman CH(1976) Amoxycillin: A new veterinary penicillin.Veterinary Record 99, 84-5

Percy DH, Black WD (1988) Pharmacokinetics oftetracycline in the domestic rabbit followingintravenous or oral administration. CanadianTournaI of Veterinary Research 52, 5-11

Pinkel 0 [1958) The use of body surface area as acriteria of drug dosage in cancer chemotherapy.Cancer Research 18, 853-6

Porter WP, Bitar YS, Strandberg JD, Charache PC(1985) A comparison of subcutaneous andintraperitoneal oxytetracycline injection mcthodsfor control of infectious disease in rats.Laboratory Animals 19, 3-6

Porter WP, Bitar YS, Strandberg JD, Charache PC(1985) Absence of therapeutic bloodconcentrations of tetracycline in rats afteradministration in drinking water. LaboratoryAnimal Science 35, 71-5

Prescott JF, Baggot ro (1988) Antimicrobial Therapyin Veterinary Medicine. Oxford: BlackwellScientific Publications

36

Rehg JE (1980/ Cecal toxin(s) from guinea pigs withclindamycin associated colitis, neutralised byClostridium sordelli antitoxin. Infection andImmunity 27, 387-90

Rchg JE, Pakcs SP 119821Implications ofClostridium difficile and Clostridium perfringensiota toxins in experimental lincomycin-associatedcolitis in rabbits. Laboratory Animal Science 32,253-7

Rehg, JE, Lu Y-S (1981) Clostridium difficile colitisin a rabbit following antibiotic therapy forpasteurellosis. Journal of the American VeterinaryMedical Association 179, 1296-7

Reinhard MK, Hottendorf GH, Powell ED (1991)Differences in thc sensitivity of Fischer andSprague-Dawley rats to aminoglycoside toxicity.Toxicological Pathology 19, 66-71

Richard Y (1990) Choix de l'anti-infectieus chez lesrongeurs. Recueil de Medicine Veterinaire 166,367-73

Richards RK, Kueter KE (19461 Competitiveinhibition of procaine convulsions in guinea pigs.Journal of Pharmacology and ExperimentalTherapeutics 87, 42-52

Ritschel WA, Vachharajani NN, Johnson RD,Hussain AS (1992) The allometric approach forintcrspccies scaling of pharmacokineticparameters. Comparative Biochemistry andPhysiology 103C, 249-53

Roine P, Ettala T (1952) Toxicity of aureomycin toguinca pigs. Nature 169, 1014

Roinc P, Ettala T, Raitio A (1953) Effects ofaureomycin and some other antibiotics onthe guinea pig. Suomen Kemistikehti 26, 17-22

Rolf LL (1993) Lack of toxicity of high oral doses ofenrofloxacin in the guinea pig (Cavia porcellus).Contemporary Topics in Laboratory AnimalScience 34, abstract P026

Rosin E (19881 The timing of antibioticadministration for antimicrobial prophylaxis insurgery. Veterinary Surgery 17, 181

Sande MA, Johnson ML (1975) Antimicrobialtherapy of experimental endocarditis caused byStaphylococcus aureus. Journal of InfectiousDiseases 131, 367-75

Sande MA, Kapusnik-Unere JE, Mandell GL (1990a)Antimicrobial agents: General considerations. In:The Pharmacological Basis of Therapeutics 8thed. (Gilman AG, Rall TW, Nies AS, Taylor P,eds). New York: Pergamon Press, pp 1041-4

Sande MA, Kapusnik-Unere JE, Mandell G111990b)Antimicrobial agents: General considerations. In:The Pharmacological Basis of Therapeutics 8thed. Gilman AG, Rall TW, Nies AS, Taylor P,eds). New York: Pergamon Press, p 1035

Schaedler RW, Orcutt RP 119831Gastrointestinalmicroflora. In: The Mouse in Biomedical ResearchVol III (Foster H, Small D, Fox J, eds). New York.Academic Press, p 339

Schatzmann HJ, Tscharner C, Tschabold M (19771Schiidliche wirkung einer oral en behandlung von

Morris

kaninchen mit ampicillin. Schweizer Archive !iirTierheikunde 119, 149-53

Smith CS, Poutsaika JW, Schreider EC (19731Problems in predicting drug effects across specieslines. The Journal of International MedicalResearch 1, 489-503

Schroder C, Matthes S, Loliger H ch. (1982)Untersuchungen iiber die vcrtriiglichkiet oralcrantibiotikamedikation beim kaninchen.Kleintierpraxis 27, 221-68

Small JD (1968) Fatal enterocolitis in hamstersgiven lincomycin hydrochloride. LaboratoryAnimal Care 18, 411-20

Taylor NS, Bartlett JG (1980) Binding ofClostridium difficile cytotoxin and vancomycinby anion-cxchange resins. Journal of InfectiousDiseases 141, 92-7

Thilstead JP 11981) Fatal diaorrhoea in rabbitsresulting from the feeding of antibioticcontaminated feed. Journal of the AmericanVeterinary Medical Association 179, 360-1

Van Miert AS (1989) Extrapolation of pharmacologicaland toxicological data based on metabolic weight.Archives of Experimental Veterinary Medicine,Leipzig 43 Suppl., 481-8

ten Voorde G, Broeze J, Hartman EG, van Gogh H119901The influence of the injection site on thebioavailability of ampicillin and amoxycillin inbeagles. The Veterinary Quarterly 12, 73-9

VPD (1991) Veterinary Pharmaceuticals andBiologicals. Lenexa, Kansas: Veterinary MedicinePublishing Company

Walsh TJ, Bacher J, Pizzo PA 11988) Chronicsilastic central venous catheterisation forinduction, maintenance and support of persistentgranulocytopenia in rabbits. Laboratory AnimalScience 38,467-71

Watson ADJ (1992/ Bioavailability andbioequivalence of drug formulations in smallanimals. Journal of Veterinary Pharmacology andTherapeutics 15, 151-9

Waynforth B (19891 Standards of surgery forrodents: Do we need to change? Scandinavian Journalof Laboratory Animal Science 16 [Suppl. 1), 43-6

Weersink A, Visser M, Vos A et al.(1991) Amoxycillin-clavulanate prophylaxis against wound infectionsafter clean contaminated surgery. EuropeanJournal of Surgery 57, 271-5

Wenzel RP (1992) Preoperative antibiotic prophylaxis.New England Journal of Medicine 326, 337-9

Wiegersma N, Janscn G, van de Waaiji D (1982)Effect of twelve antimicrobial drugs on thecolonisation resistance of the digestive tract ofmice and on endogenous potentially pathogenicbacteria. Journal of Hygiene (Cambridge) 88,221-30

Young JD, Hurst WJ, White WJ, Lang C Max (19871An evaluation of ampicillin pharmacokinetics andtoxicity in guinea pigs. Laboratory AnimalScience 37, 652-6 [with ERRATA 38,345]

antibiotic therapeutics in laboratory animals

Documents