Ceftazidime, a third generation cephalosporin, is active againstGram negative bacteria, as Enterobacteriaceae (Escherichia coli,Proteusspp.,Klebsiellaspp.)andPseudomonasaeruginosa,andsomeGram positive cocci as staphylococci and streptococci. Modifica-tionsinthepharmacokineticsofantimicrobialagentsraiseamajorconcern as successful antimicrobial treatment depends on appro-priate plasma and tissue concentrations. Previous studies havedetermined that, for time-dependent killing antimicrobials, suchas beta-lactams, the major determinant for bactericidal activity isthe time for which serum drug concentrations remain above theminimum inhibitory concentration (T > MIC) for the offendingpathogen (Craig, 1998; Drusano, 2004; McKellar et al., 2004).Occasionally, parenteral administrations are considered inter-changeable by the veterinary practitioner. However, it is knownthat several factors, including route of administration, may alterthe rate and extent of drug absorption and disposition. We haverecentlyprovedthelackofinterchangeabilityoftheintramuscular(i.m.) and subcutaneous (s.c.) routes of administration forcephalexin in cows (Waxman et al., 2008).Ceftazidime is commercially available for intravenous (i.v.),i.m. and s.c. administration. Ceftazidime pharmacokinetics hasbeen well characterized in cats (Albarellos et al., 2008), calves(Soback & Ziv, 1989), cows (Rule et al., 1996), sheep (Rule et al.,1991) and dogs (Matsui et al., 1984; Kita et al., 1992; Mooreet al., 2000). However, to our knowledge, no studies have beenconducted to assess T > MIC calculated after the differentparenteral administrations in dogs. This study compares thepharmacokinetic behaviour and the efficacy predictor T > MICagainst P. aeruginosa of a single dose ceftazidime following i.v.,i.m. and s.c. administration to dogs. The aim was to determinewhether these administrations would provide T > MIC valuesthat would suggest similar clinical outcomes.Six female beagle adult dogs (body weight 15.61 ± 5.47 kg)were included in this study. All dogs were in good health, asdetermined by history, physical examination, haematologicaland biochemical tests and urinalysis. None of the dogs had beentreated with antibiotics in the previous 2 months or had ahistory of allergy to beta-lactams. The protocol was approved bythe Institutional Animal Care and Use Committee of theVeterinary Science School, University of Buenos Aires.Single doses of an aqueous solution of ceftazidime pentahy-drate (Ceftazidima Richet, Laboratorios Richet S.A., BuenosAires, Argentina) were administered by the cephalic vein(20 mg⁄kg), and by the i.m. (25 mg⁄kg) and s.c. (25 mg⁄kg)routes. Each dog received the three treatments following acrossover design with a 2-week washout period. Blood samples(2 mL) were collected from the jugular vein into heparinizedtubes at 0, 0.08, 0.16, 0.33, 0.5, 0.75, 1, 1.5, 2, 2.5, 3, 4, 6, 8,10, 12 and 24 h after ceftazidime administration. Samples wereimmediately centrifuged and the plasma was stored at )20 Cuntil assayed.Ceftazidime plasma concentrations were determined in tripli-cate by a microbiological assay (Bennet et al., 1966) usingEscherichia coli ATCC 25922 as test organism. Standard curves ofceftazidime were prepared in pooled canine plasma and runsimultaneously with test samples. The quantification limit was3.125 lg⁄mL. The correlation coefficient for the regression lineof the standard solution was 0.98. The within and between-daycoefficients of variation and the accuracy (bias) of the assay were<2.5%, 10% and 5%, respectively, in the range of observedconcentrations (from 3.125 to 200 lg⁄mL).A computer program (