‘ d. a: .3 .9. a, 1..“ L. .1 s29. .9€.h...>!.:,fl (- cl. . h. .c ill .1? .r... Eu: ‘ i..!..v.x.3 s.. :1 LLnti... 10.37.: 51!... :r a}... . llilyi.\-K.r . h Jaws. b. no V1.8... .§-.,T 7..» VP! 31F»,€Y. nu, hr Jw :1 tk v. e.) t. J”. . ‘ . .K. . .77 .n. ,4?! Nil-vi! <"'-3,- a .3. . ‘I vt’i ‘ ‘.| :1... 15.2515... 1; o. . E w 1).... .t: > :1 v V ‘1'? if}: :3. z .5: I ..r\ A ‘ L 1. I. . ,u! .rb.v..vp..‘ “nth .. tuna». up: . ‘ 7 ‘ , . . . {lite . . a. :a: , 3:... ..... . , :3. ‘Ir....: .. . ..l 3: e113. . \u. . a a. be . t... 4. ‘od fidwflf. $.32. 1-. «Brawn m5 .un......_...h.e..... , ...s,..xo....~mua....i ..-..l ‘ 1311.12”. tr, ‘ “In I ‘ n..mnmeu§.wfim§m 111111111111111111111111111111111111 1111111 111111111 ”IL 31293 00789 98541 ‘ LIBRARY 1 Michigan State University 1 L J W This is to certify that the thesis entitled Pharmacokinetic evaluation of ceftiofur sodium in serum, tissue chamber fluid, and bronchial secretions from beef-breed calves presented by Steven Leroy Halstead has been accepted towards fulfillment of the requirements for Masters degree in Science M/‘zér/ Major professor Date _.9_AugusL_1220_ 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution - . .f"’ /,. -, z" j v’ PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE ll iii MSU Is An Affirmative Action/Equal Opportunity Institution ‘ ammuna-pd PHARMACOKINETIC EVALUATION OF CEFTIOFUR SODIUM IN SERUM, TISSUE CHAMBER FLUID, AND BRONCHIAL SECRETIONS FROM BEEF-BREED CALVES BY Steven Leroy Halstead A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Large Animal Clinical Sciences 1 990 Cefi infec Orga phar adml and t micro Naxce abSOr inCrea elimjné associ. \ é ¢ 128.0 Non-BRO assoc. organismp No, isolates M990 (uglm!) Escherichia ELI 10 0.25 Salmonella spp 6 2.0 Staphylococcus aureus 7 32.0 Streptococcus spp 17 0.06-2.0 This product is currently approved for use at 1.1 mg/kg body weight, given intramuscularly. Treatment of calves with twice this dose produces serum and lung tissue homogenate concentrations in excess of the in vitro MIC concentration for target organisms“. Equivalent data for the label-recommended, FDA approved dose has not been generated, nor has data for serum or tissues by means other than tissue homogenation (tissue fluid data) for various doses been generated or published. 1‘ Miller CC. Ceftiofur research and development. Personal communication. QUANTIFICATION OF DRUG ACTIVITY As stated earlier, understanding the disposition of a drug at secondary compartment locations, the most common location of pathogenesis by infective organisms, is critical to the making of valid therapeutic decisions. To quote Weinstein et al: 'While information concerning the concentration of drug in the blood is of great importance in studying it’s absorption and excretion, it may be of little or no value in ascertaining the actual quantity of antibacterial substance at the site of infection. There is little doubt that the tissue level of penicillin is related to the blood concentration, but the degree of relationship, particularly as it involves the duration of antibacterial effect, is not settled.’ (Weinstein et al., 1951). Measurement of kinetics at these peripheral locations is rarely as straightforward as in the central compartment. Collection of samples that are representative of conditions in situ and that can be quantitatively assessed without bias is simple when blood is the tissue, but becomes much more difficult for other tissues. Serum Sampling Techniques Serum pharmacokinetic parameters for many drugs have been reported. Whether this data is the main subject or associated with data for other tissues, the technique of sample collection in most cases involved simple percutaneous aspiration of whole blood via vascular puncture, separation of the fluid portion from the solid portion by centrifugation, filtration, or innate clot formation, followed by processing the fluid portion in a specific biologic or chromatographic assay. 35 Tissue Sampling Techniques Collection of representative samples from other locations may be as straightforward as blood or serum collection; cerebrospinal fluid, bile, and urine are all easily collected as means of evaluation of the central nervous system, liver, or urinary system, respectively. Fluids from most other tissues, however, are not easily collected in volumes sufficient for pharmacokinetic analysis by current techniques. The fluids providing nutrition, lubrication and protection for these tissues exist in small quantities dispersed across large surface areas or potential spaces that are currently impossible to access directly. The Challenge, then, is to collect these fluids as pure, representative aloquots of sufficient quantity. To this end, many techniques have been reported. Homogenation of tissue is one commonly used technique. Tissue homogenates contain blood, lymph, interstitial fluid, cytoplasm, and possibly other fluid components such as synoviai fluid or glandular secretions (Bergen, 1981). Any of these may be present in varying quantity depending on the time and technique of collection and the processing methodology, and may contain substantial amounts of the compound of interest. Such non-quantified sources may contribute significamly to the total sample compound measurement, leading to false interpretations of true tissue fluid concentration of that compound. The possibility also exists that enzymes may be present in homogenates that are capable of deactivating the component being assayed, reducing the measured concentration only when the tissue has been unnaturally altered by the homogenation procedure. Alternatively, protease enzymes may be released by homogenation, altering quantification accuracy by releasing drug that was protein bound , and therefore not biologically active in the intact tissue (Bergen, 1981). Quantification of drug concentration cannot be performed accurately, unless the contribution of drug from each fluid component can be determined without alteration by other substances. Methods exist for adjusting the measured values relative to blood contributions, based on measured serum concentration, tissue water volume, and blood volume; radiolabeled erythrocytes; hemoglobin 36 content (Bergen, 1981), or hemetocrit (Wise, 1986). Each of these methods is subject to miscalculetion, which, combined with error from failing to account for unmeasurable drug contributions or alteration due to technique (Persons et al., 1976) may result in significant misrepresentation of drug concentration. Measurement of drug concentration in lymph collected from common, central lymphatics or from regional, specific tissue lymph ducts has been reported (Bergen, 1981). Central and thoracic duct lymph are poor choices if isolated target tissue data is required because they collect drainage from many tissue beds. Differences exist between thoracic duct, central. and peripheral lymph for common substances such as protein and lipids, making the report of dmg concentrations based on any one of these sources suspect to much scrutiny because of the variability each displays in what it transports, and also due to effects on drug concentration of these transported molecules (Bergen, 1981). Lymphatic vascular endothelium is not passively permeable to all interstitial fluid components, and antibiotics, especially those bound to albumin, may not pass predictably or reliably into the lymphatic fluid (Bergen, 1981). Therefore, even though lymph might in some cases be collected easily and without contamination by blood, intracellular fluid or specific secretions, it is not an ideal fluid for tissue-dmg concentration evaluation (Bergen, 1981). Another method of tissue fluid collection is the skin blister, induced by suction or contact with an irritant such as cantharidin (Bergen, 1981; Wise, 1980). The fluid enclosed by blisters is in communication with intravascular fluid across the non-inflamed vascular membranes, and is assumed to represent tissue fluid (Bergen, 1981). The irritant-induced blister is less consistent than suction blisters in size and level of deep inflammation, and thus results in less consistent drug measurements. This technique is not readily applicable in studies of pharmacokinetics on large animals due to the potential for damage and rupture of the blisters during routine handling. 37 Other methods of tissue fluid collection such as subcutaneously implanted threads or fibrin clots have been used (Hoffstedt and Walder, 1981; Bergen, 1981) but are not favored by many investigators because they lack a counterpart in natural disease (Bergen and Weinstein, 1974). Sterile chambers (tissue chambers) are used frequently as sources of tissue fluid. These chambers have been constructed from teflon, wire mesh, steel spring, polypropylene, thermoplastic and silicone rubber (Chisholm et al., 1973; Clarke et al., 1989a; Higgins and Lees, 1984; Walker et al., 1989; Walker et al., 1990). Surgically implanted in the subcutaneous space, they are believed to be continuous with interstitial space and interstitial fluid. Studies comparing acid-base, pH and solute concentration between tissue chamber fluid, tissues and local lymph (Bergen, 1981), and between tissue chamber fluid and blood (Clarke et al., 1989b) support this claim, and radioisotope-labelled solutes has shown that marker penetration into chambers occurs rapidly, producing identical concentrations as those in lymph (Bergen, 1981). Regardless of the structural material composing the tissue chamber, the principal of operation is the same. After implantation the cage fills with blood, serum and tissue fluid. As organization of this fluid progresses, the chamber becomes lined and surrounded with fibroblasts, capillaries and other interstitial tissue components. Proliferation of these tissues will eventually fill the chamber lumen completely, as the perforations that allow passage of fluid also expose the lumen to the ingrowth of this new interstitial tissue. The rate of fluid diffusion and therefore the time until equilibrium is reached between tissue fluid and Chamber fluid is determined by the proportion of perforated surface area to total surface area. Diffusion rates vary for specific solutes, larger molecules such as albumin requiring longer time periods than smaller molecules (Chisholm, 1973). This will be discussed further below. Use of the chambers should be delayed until at least 4 weeks after implantation to allow the acute inflammatory phase to subside. This can be demonstrated by stabilization of tissue chamber 38 fluid erythrocyte and leucocyte numbers and solute content at 4 weeks compared to fluid collected shortly after chamber implantation (Clarke et al., 1989b). Some tissue fluids for which phennacokinetic data is desired can be accessed directly, but the convenient collection of these fluids in sufficient quantity is difficult. This is frequemly the situation with serosel or mucosal surface fluids, or surgical wound fluids. Absorption of the fluid into a suitable inert media that can be used directly in the quantification assay, or from which the agent of interest can be extracted for assay, has been found to be acceptable (Wise et al., 1981; Hajer et al., 1988). This technique involves placing a pro-weighed, desiccated absorbent device on the epithelial or wound surface for fluid collection. The common medium for absorption are 6.35 mm (0.25 inch) diameter sterile filter paper antibiotic sensitivity assay discs. After collection, the discs are re-weighed to measure the fluid volume collected, then used directly in the biological assay. This technique requires adjustment of the final assay- determined antimicrobial concentration for variation in the amount of fluid absorbed, as volumes larger or smaller than the volume of the standard from which the concentration is derived will result in correspondingly larger or smaller sample concentrations. Depending on the interests of the investigators, samples may be rejected if blood contamination occurs or exceeds some arbitrary value. Swabs have been used as an alternative to the disc as the medium of secretion collection in this technique (Hajer at al. 1988). Similar adjustments for fluid volume variation were made as are required with disc absorption. These investigators also chose to bypass the upper airway and oropharynx entirely by installing tracheostomy devices, a technique that might produce alterations in the characteristics of the respiratory mucosa due to bypass of the normal hydration and warming mechanisms (Murray, 1986). 39 Other investigators have used fluids as the collection media (Mair, 1987; Slocombe et al., 1985). Washing the pulmonary mucosa with a physiologically benign solution and measuring the amount of drug in the effluent produces acceptable results (Slocombe et al., 1985) but requires adjustment for dilution based on the differential of some endogenous compound in the animal's serum and in the collected effluent. This technique may not be accepted well by normal, alert subjects, requiring tranquiilzation that could alter cardiovascular function as well as other physiologic parameters which in turn may affect the final results. Quantification Techniques Samples, once collected, require processing and quantification of drug content; in the interest of toxicity, metabolic activity, or oral dosing compliance, for example, or for pharmacokinetic evaluation; as is the focus of this discussion. One of two basic assay techniques are commonly used: high pressure liquid chromatography (HPLC) or microbiological assay (Bioassay). High pressure liquid chromatography techniques require technological expertise and sophisticated equipment, but are extremely accurate, reliable, and relatively rapid, delivering results in 30—60 minutes. Chromotagraphic systems can measure original compounds, metabolites, or both, and they report all drug present in the sample. Bioassays are labor and materials intensive, sensitive to day to day variation, highly subject to technical error, less sensitive, and less specific than HPLC (Edberg, 1986). Bioassays, however, measure only biologically active compounds, whether those compounds are parent compound or metabolites. For this reason, bioassay techniques are still in common usage in pharmacokinetic studies (T oothaker et al. 1987). HIGH PRESSURE LIQUID CHROMATOGRAPHY Analysis of many water soluble compounds, including penicillins and cephalosporins, is frequently carried out by HPLC, and is reported to 40 be used preferentially in the analysis of third generation cephalosporin in secondary compartments (T oothaker et al., 1987). The technique requires extraction of the dmg with a specific solvent, separation of the drug onto solid phase by HPLC, detection of the effluent off the solid phase by spectrometry, and quantification of the amount of antimicrobial present by peak height or peak area analysis. HPLC techniques are commonly used in labs to measure anti-convuisant and anti-arrhythmic drugs, or any drug that has a narrow toxic/therapeutic ratio and requires close monitoring, as in the case of antimicrobials that have high potential for nephrotoxicity. While these concerns exist for a small number of antimicrobials, the short period of time required for processing and the competitive cost relative to other techniques makes HPLC analysis attractive (Edberg, 1986). MBIOJIQGICAL ASSAY Ten years ago, 75-85% of clinical laboratory assays were microbiological in design. By 1986, only 30-40% of the antimicrobial studies conducted were microbiological assays, with the difference being made up by non-isotopic studies such as immunoassays and HPLCs (Edberg, 1986). Nonetheless, microbiological assays are the cornerstone method as techniques for pharmaceutical assessment, with the advantages of low cost, simplicity and biological significance in that actual antimicrobial activity is the basis for the measured result (Edberg, 1986). Bioassays can take the form of agar diffusion techniques, which will be discussed further, or turbidimetric techniques, where the concentration of an antimicrobial agent is determined in comparison to microbial inhibition by known concentrations, either by direct photometric measurement or indirectly by pH change as the indicator of substrate metabolism by growing bacteria (Bany, 1986). Agar diffusion assays utilize the development of zones of bacterial growth inhibition to determine the amount of antimicrobial agent present. This zone is formed when the amount a 1_ - 7. glib.- 41 of diffusing antimicrobial drug just capable of inhibiting microbial growth (the critical concentration) reaches a density of organisms larger than can be suppressed. When bacteria are seeded into or streaked onto agar media and placed in a favorable environment they begin to utilize the nutrients in that media for growth and reproduction. Agar media is a complex and not fully understood gel of polysaccharides and cations (calcium, magnesium, copper, zinc, and iron) dissolved in water (Barry, 1986). As these nutrients are used by the bacteria, a concentration gradient develops between the depleted area of growth and surrounding areas of the gel. Nutrients move across this gradient by the laws of diffusion, as do bacterial waste products, some of which may have a protective effect for the organism (Barry, 1986). If a spot source of antimicrobial is placed in this system it behaves similarly, diffusing along a concentration gradient in accordance with these same diffusion principles. Thus, a dynamic system develops with bacteria colonies appearing in regions where nutrient needs are met and antimicrobial concentration is below levels that inhibit their growth. In regions of sufficient antimicrobial concentration, the growth of the organisms will be suppressed, and the agar will remain clear of visible bacterial colonies. Zone formation depends on the properties of the system, and varies as those conditions change. These properties include such antimicrobial properties as molecular concentration, size, shape, and charge; and media properties such as viscosity, temperature, moisture content, and ion content. The critical concentration of drug is given by the formula (Barry, 1986): In m' = in rn0 (xz/4DTO) Where: In m' = natural log of the critical concentration In m0 natural log of concentration of drug applied to agar surface (point source) 42 x2 = square of distance between edge of point source (antimicrobial reservoir) and edge of the zone of inhibition D = diffusion coefficient of antimicrobial for the particular test system conditions To = critical time at which zone position is determined A straight line relationship should develop if in m0 is plotted against x2 with different initial concentrations run on identical test plates under identical conditions (all other factors are held constant). The antimicrobial agent may be introduced to the system in various ways. Wells may be cut into the agar and the drug instilled with pipettes, or stainless steel cylinders may be placed on the surface and filled with the agent. Another method involves using filter paper discs, placed in full contact on the media surface, as reservoirs. All serve as point sources of antimicrobial drug, from which diffusion freely occurs, although one source reports the well technique to be 5-6 times more sensitive than the filter paper disc technique (Sabath et al., 1969). The zone of inhibition will form at a distance (x) after a certain time period. This period is known as the critical time (T 0) after addition of antimicrobial agent to the system and initiation of incubation. it is Independent of the concentration of the drug in the system but does reflect the rate of microbial growth at the time when the antimicrobial compound is diffusing most rapidly through the media (Barry, 1986). Therefore, it is the time required for the indicator organism to reach a critical cell mass that exceeds the inhibitory potential of the antimicrobial agent at the concentration present some distance x from the drug reservoir (Barry, 1986). it is expressed by the relationship: To - h = x2/4D ln(mo/m’) Where: 43 To = critical time h = hours of pro-incubation x2 = zone size squared D -= diffusion coefficient of antimicrobial agent ln(mo/m’) a natural log of concentration of drug in reservoir divided by the critical concentration From the critical population formula: To = L + G Iogz(N’/N0) To = critical time L = lag time G = generation time N’ - critical population at critical time (to) N0 = number of viable cells at time = 0 Gnoculum density) It can be seen that the initial inoculum and the generation time are most important in determining the critical time, and therefore the size of the zone of inhibition. For example, longer generation times or more dilute initial inoculates will give longer time periods until critical population is reached, and correspondingly larger zone sizes because of longer periods for antimicrobial diffusion before critical populations are encountered. Generation times are intrinsic properties of the organism that are dependent on nutrient availability and incubation temperature. Identical growth conditions will produce identical generation times, an important factor in reproducibility of bloassay data. 44 MINIMUM INHIBITORY CONCENTRATION Pharmacokinetic information is useful only if it can be compared to some objective reference. With antiarrthymic drugs, for example, this comparison parameter might be the concentration of drug required at a particular receptor to block spontaneous excitation, thus preventing arrthymias. Antimicrobial pharmacokinetic data is reported relative to the minimum concentration of antimicrobial drug required to inhibit growth of target microorganisms (MIC), or the concentration required to ineversibly inhibit (kill) an inoculum after a defined incubation period (Minimum Bactericidal Concentration, MBC) (Thrupp, 1986; Vogeiman et al., 1988). While this comparison is most appropriate for beta-lactam compounds, antimicrobial efficiency of other classes of compounds, aminoglycosides, for example, are best defined by comparison of log area under the curve (Vogeiman et al., 1988). The efficiency by which a beta-lactam compound exceeds the MIC is determined by a combination of the rate at which it penetrates the bacterial outer membrane, the resistance of the compound to various B-lactamase enzymes, and the kinetic parameters of the interaction of the compound with penicillin-binding proteins (Spratt and Cromie, 1988). Knowing the concentration of drug required to inhibit growth or kill a known or suspected pathogen allows calculation of the therapeutic ratio, which is equal to serum or tissue fluid concentration divided by the MIC (Sande and Mandell, 1985). This value provides an index value for efficacy in clinical cases of bacterial infection, where antimicrobial drugs that produce therapeutic ratios of 4 or greater are considered to be appropriate choices (Sande and Mandell, 1985). Techniques for determination of MlC's involve 2-fold dilutions of the antimicrobial agent either in suspensions of broth or solid (agar) media (Thrupp, 1986). Standardized inoculates are exposed to the media and incubated. The endpoint is read as the least concentration of drug by which the organism’s growth is inhibited. The actual inhibitory concentration is between this concentration and the next lower concentration in the dilution series (Thrupp, 1986). Standards for procedures have been published‘. 1 National committee for clinical laboratory standards Tentative standard: M7-T2. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. 2nd ed. Villanova, Pa 1988. OBJECTIVES OF STUDY The objectives of this study were twofold. First, to determine the pharmacokinetics of ceftiofur sodium in semm, tissue chamber fluid (TCF), and bronchial secretions (BRS) collected from feeder-age, beef-breed calves following the first and the fourth doses of once—daily intramuscular injections at 1.1, 2.2, and 4.4 mg/kg. Second, to measure in vitro the minimum concentration of ceftiofur sodium required to inhibit the growth of various bacterial isolates recovered from cattle. A comparison of these MIC values and pharmacokinetic parameters at the different tissue locations could provide guidelines for ceftiofur sodium antimicrobial therapy. 45 MATERIALS AND METHODS Experimental design This study involved trials with three dosages of the antimicrobial agent ceftiofur sodium. Each trial consisted of 4 days of single daily intramuscular drug injections, with at least 4 weeks between dosage trials. Serum, tissue chamber and bronchial secretion samples were collected from each call at time intervals after the first dose and after the fourth dose. All 4 calves were treated with the first and fourth dose immediately after time zero samples were collected, for all 3 dosages. Drug administration to individual calves was staggered by 10 minutes on sampling days to allow time for sample collection. Samples were then processed by bioassay for pharmacokinetic evaluation. Animals Four beef-breed (Angus cross) calves, 2 male and 2 female (all sexually intact), were used in this study. The calves were purchased from a local commercial beef hard at approximately 4 months of age, and transported to the Veterinary Research Farm at Michigan State University, East Lansing, Michigan. The calves were maintained under natural environmental conditions on pasture grass (timothy, orchard grass, and other common grasses in natural growth), supplemental alfalfa hay, ground corn, and fresh water from an automatic waterer. 46 47 News Chambers Six cm lengths of silastic tubing3 (1 cm outside diameter, 8.5 mm inside diameter) was used to make tissue chambers as described by Chisholm (1973). Ten circular holes of 2 mm diameter were cut in the tubing near each end, and the ends closed using silicone sealantd. The perforations resulted in approximately 20% of the surface area of the cylinder being open. The chambers were gas sterilized with ethylene oxide" and aerated for 24 hours prior to surgical implantation. Surgical implantation Of Tissue Chambers The calves were in the range of 300350 pounds (136-160 kg) body weight when the silastic tubing tissue chambers were implanted. The calves were placed in a restraint chute, sedated with approximately 0.10 mg/kg xylazine lVl, both sides of the neck were prepared for surgery, and anesthetized locally with 5 ml Infusions of 2% lidocainem in the area of the skin incisions. Four horizontal incisions were made on each side of the neck, approximately 5 cm in length and located midway between the jugular furrow and the dorsal extent of the neck (ligamentum nuchae). The incisions were spaced evenly on a line drawn from the poll of the head to the glenoid area of the scapula. Subcutaneous pockets were made through these incisions in dorsal and ventral directions by introducing scissors and bluntly dissecting the subcutaneous tissue. Sterile tissue chambers were placed in these pockets, one dorsal and one ventral, with their long axis aligned dorsoventrally. After tissue chamber placement the skin was closed with number 2 vetafil in a simple continuous pattern. Seven chambers were implanted on each side of the neck. The calves were then placed on ceftiofur at 1.1 mg/kg, once daily, IM, for five 3 Dow Corning, Midland, Michigan. “ Steri-Vac 4000, 3M Corp. St. Paul, Minnesota. H RompunR, Haver, Bayvet div., Shawnee, Kansas. “' Butler 60., Columbus, Ohio. 48 days to prevent post-operative infections. Sutures were removed after ten days, and the chambers left undisturbed for 4 weeks from the day of implantation. The course of the experiment ran for the next eight months, during which time the calves gained approximately 400 pounds (182 kg) each, and reached approximately 11 months of age. Samples The samples were collected according to the following schedule: Serum and tissue chamber fluid (TCF) samples collected at time (0 = 0, 0.5, 1.0, 1.5, 2, 3, 4, 5, 6, 8, 10, and 12 hours after dmg administration. Bronchial secretion samples (BRS) collected at T = 0, 1, 2, 3, 4, 5, 6, 8, 10, and 12 hours. Sample collection and processing was conducted 2 times for each dosage level: after the first dose and after the fourth dose, to generate both single and multiple dose data. This investigation involved once-daily injections of Naxcel“ at three dosage levels: 1.1 mg/kg (label recommended dose), 2.2 mg/kg, and 4.4 mg/kg, administered by deep injection into the gluteai muscles using 18 gauge, 3.81 cm long sterile needles. Successive daily injections were alternated between left and right sides of the calves. The drug used for this experiment was the commercially available Upjohn product, reconstituted and stored according to the label recommendations. Sample Collection The calves were brought to the Veterinary Clinical Center at Michigan State University and housed in box stalls 48 hours prior to sample collection to allow adjustment to the indoor environment and for drug trial preparation. The hair was clipped over the neck and jugular veins, the area was scrubbed with an iodine-based detergent“, and the animals were weighed. During the sample collection period the calves were tied loosely with rope halters, from which were released periodically for access to food and water. “ Betadine surgical scrub, Purdue Fredrick Co., Norwalk, Connecticut. 49 SM In the first sample collection period (1.1 mg/kg dosage) the venous blood samples were collected by 20 gauge, 3.81 cm needle venipuncture with aspiration into 10 ml vacutainer clot tubes°. For the second and third collection periods (2.2 and 4.4 mg/kg dosages), jugular venous blood samples were collected through 5-1/4 inch, 16 gauge cathetersp that were placed the day before the sampling period began. This change was instituted to reduce discomfort and resistance in the calves. The catheters were sutured to the skin and wrapped in elasticonP for protection, and flushed with 5 mls of heparinq prior to use. During collection periods the catheters were flushed with heparinized saline (10 mls heparin at 1000 units per ml in 1 liter of commercial sterile saline solution) between sample collections to prevent occlusion by clotted blood. The heparinized saline and first 5 mls of blood were discarded before blood was collected to be saved for assay. Blood was pulled from the catheter by 12 mi syringe, then transferred to 10 mi vacutainer clot tubes. (5 to 8 mls of blood were collected regardless of technique.) Five mls of undiluted heparin (1000 units/ml) were placed in the catheters for overnight periods. TISSUE CHAMBER FLUID Tlssue chamber fluids for all sampling periods were collected into 3 ml syringes by direct percutaneous puncture of the tissue chambers with 22 gauge, 1 inch needles. Tlssue chamber fluid was then transferred to 3 ml vacutainer clot tubes. Tissue chambers were only used once per sampling period (dosage level), with the most cranial chamber on the calf’s left side being used first, progressing ventrally then caudally, while alternating sides. This pattern was followed throughout all sampling periods. ° Becton-Dickinson, Rutherford, New Jersey. P Johnson and Johnson, New Brunswick, New Jersey. ‘1 Lympho med, inc., Melrose Park, illinois. 1,000 units/ml. 50 BRONCHIAL SECRETIONS The bronchial secretion samples for the first sampling period were collected by passing a 1.2 cm outside diameter equine nasogastric tube through the calf’s nostril, ventral nasal meatus, larynx, trachea and mainstem bronchi until it lodged in the mainstem bronchi to either of the caudal (diaphragmatic) lung lobes. No attempt was made to determine the specific location at which the tube lodged within the lung. The tube lumen was then swabbed with a small plug of absorbent cotton held in the jaws of endoscopic biopsy forceps’. The swab was withdrawn, and a preweighed sterile disc“ was clamped in the biopsy jaws. The disc was passed through the tube, beyond the lodged and until it encountered resistance to further passage. The disc was left in contact with the mucosa for 20 seconds, withdrawn, and placed in a sterile 10 ml vacutainer clot tube. This procedure was repeated for a total of 3 discs per sampling time. This technique were modified slightly for the 2.2 mg/kg and 4.4 mg/kg sampling periods to minimize discomfort experienced by the calves. instead of passing the tube through the nostril and nasal passage, a wooden mouth gag was placed and 8 Cole endotracheal tubet was passed through the gag and into the larynx. The equine nasogastric tube was passed through the Cole tube and samples collected as previously described. Sample Processing Bronchial secretion and tissue Chamber fluid samples were placed in an ice bath immediately after collection, where they remained until further processing. Jugular blood was allowed to clot at room temperature for 1 hour, then held in the ice bath until processing. ‘ Olympus Medical instruments Div., Lake Success, New York. 3 No. 1599-35, Difco Laboratories, Inc., Detroit, Michigan. ‘ Lane Mfg., Denver, Colorado. 51 Serum samples and tissue Chamber fluid samples were centrifuged at 1,0009 for 10 minutes at 4 C. Semm samples were separated with assistance from plastic centrifugation beads“. After centrifugation the serum and tissue chamber fluid samples were transferred by glass Pasteur pipette to 1.2 ml Nunc freezer vials". Bronchial secretion samples were weighed as a composite of the 3 discs collected per sampling time for determination of the volume of absorbed secretions. Any grossly visible mucous or other debris were removed prior to weighing. All samples were kept at 4 C until assayed, then transferred to -70 C for long term storage. Bloassay PLATE PREPARATION The bioassay utilized two layers of media in standard 150 x 15 mm petrl plates”. The top, or seed layer, contained the indicator organism in a thin layer of media. The bottom, or base layer, served as a buffer for diffusion of the samples after they had passed through the seed layer. This prevented the sample fluid from being trapped by the bottom of the plate and reflected to the seed layer, potentially affecting the size of the zone of inhibition. The base plates were prepared by pumpingx a base layer of 30 mls of antibiotic test media number 11y into the petrl plates. The media was prepared as per manufacturer‘s rehydration and autoclaving recommendations, with 1 drop of PouriteR2 per liter of media to reduce bubble formation on the plates. The plates were allowed to cool before moving or stacking to prevent disturbance of the base layer. Once all base plates had cooled sufficiently to solidify, they “ OTl Specialties, Santa Monica, Callfomia " lntenned, Denmark. " American Scientific Products, McGraw Park, illinois. ‘ Technomara media pump, Zurich, Switzerland. 3' Difco Laboratories Inc., Detroit, Michigan. ‘ Analytical Products, Inc., Belmont, California. 52 were moved to a large walk-in refrigerator for storage. Base plates that were not used within two weeks were discarded. The seed layers were prepared the day prior to sample collection using the same precautions as used with the base layers to prevent irregularities in the media. The seed layer consisted of 20 mls of Antibiotic Test Media 1 with Providencia alcalifaciens ATCC 9886“ as the indicator organism. The E. alcalifaciens inoculum was prepared by growing the organism, taken from minus 70 C storage, overnight on blood agar plates at 37 C in a 5% CO2 incubator. The blood agar plates consisted of 5% defibrinated sheep’s blood“, 1% Yeast extract“, and 1% heat inactivated horse serumdd in Tripticase Soy Agar“. Three to five colonies of the organism were then transferred to 200 mls Brain-Heart Infusion broth" and incubated in a shaking water bath at 80 rpm for 5 hours at 37 C. After incubation the inoculate was diluted to produce a standard optical density of 0.042 at a wavelength of 650 nm“, which corresponded to a McFarland density of 0.5 (108 colony forming units/mi). Three mls of this inoculum was added per liter of seed media (0.3% inoculum) after the antibiotic test media 1 had been held at 120 C for 15 minutes and allowed to cool to 42 C. PouriteR was also included in the seed media. Dispensation of the seed layers began after allowing 2-3 minutes for media and inoculum to mix by rotating paddle agitation in the media preparator“. The seed media was pumpedhh “ American Type Culture Collection, Rockville, Maryland. 1’" Cleveland Scientific, Bath, Ohio. °° BBL division of Becton-Dickinson, Cockeysville, Maryland. dd Gibco Laboratories, Lawrence, Massachusetts. °° Difco Laboratories Inc., Detroit, Michigan. “ Model J-A spectrophotometer, Coleman Scientific. 3‘ Model SH110 Jouan, Winchester, Virginia Technomara media pump, Zurich, Switzerland. 53 onto the base layers after they had been pro-warmed for 2-4 hours at 37 C. This pro-weaning was used to prevent rapid gelling and possible seed layer irregularities. The plates were spread out and allowed approximately 1 hour to cool to room temperature after pouring, then stored at 4-5 C until use. Fresh plates were prepared for each sample day, as they were found to be unreliable if stored for more than 48 hours. STANDARD CURVE GENERATION The standard curve (reference curve) was generated for this bioassay system using dilutions of 5.0, 2.5, 1.25, 0.625 (midpoint concentration), 0.3125, 0.156, and 0.078 jig/ml of Ceftiofur sodium standard“. The standard was assayed by The Upjohn Company at 895 micrograms ceftiofur per milligram of total compound, or 1,000 pg in 1.12 mg of the standard. Thus, the standard curve dilutions were made by dissolving 11.2 mg of standard in 10 mls bovine serum“ (heat inactivated, filter sterilized) to make a concentration of 1,000 pg/mi, from which further dilutions in the same bovine serum were made to reach the range listed above. These dilutions were plated in triplicate on the prepared bioassay plates, alternated with the midpoint dilution of 0.625 ug/ml, and each plate was reproduced to create 3 plates, or 9 standard curve zones and 9 midpoint zones. The lower limit of sensitivity of this bioassay system was 0.156 pg/ml, as determined by the straight line portion of the semi-log plot of inhibitory zone size against drug concentration (Figure 4). EXPERIMENTAL SAMPLES Serum and tissue chamber fluid were assayed by methods previously described (Walker et al., 1989). Briefly, serum and tissue chamber fluid were added to the plates as 20 pl aloquots on sterile blanks“. Each sample for each animal was tested in triplicate on each plate, and each plate was replicated for a total of three plates. The fluid 1‘ The Upjohn Company, Kalamazoo, Michigan. 33 Collected from MSU Veterinary Clinical Center blood donor cows. k“ No. 1599-35, Difco Laboratories lnc., Detroit, Michigan. 54 samples were alternated with similar discs containing ceftiofur in a standard concentration that corresponded with the midpoint dilution of the standard curve series. Thus, nine sample discs and nine standard midpoint discs with corresponding zones of inhibition were generated for each serum and tissue chamber fluid sample. The bronchial secretion samples, being absorbed directly into the sterile blanks from the mucosa, were placed directly on the seed media in the same arrangement with the standards as the fluid samples. Only one plate of three bronchial secretion samples and three standard discs was set up for each animal at each sample time. All plates were incubated overnight at 37 C under aerobic conditions. The plates were placed in stacks of no more than two for the first 2 hours, after which they were stacked to the shelf space capacity of the incubator. All plates were read manually, measuring zone sizes for the test animal samples first. The plates generated for the first dosing period (1.1 mg/kg) were read on an overhead projector with a millimeter scale overlay. The plates for the 2.2 mg/kg and 4.4 mg/kg dosing periods were read on a light box using an electronic caliper device”. All zones were read to the nearest millimeter size and recorded individually. The initial bioassay data processing was done on microcomputer using custom software from Micromath Inc”. Initial data analysis involved averaging zone sizes to reduce the data to one value for each sample type from each animal at each sample time. These averages were then adjusted for plate to plate variation using the difference of the standard midpoint values (also n Mitutoyo Corp. Tokyo, Japan. m Micromath lnc., Salt Lake City, Utah. 55 averaged) for each plate and the average value for all the standard midpoint samples on all plates. Adjusted averages were then converted to microgram per milliliter (pg/ml) values based on the standard curve. These [Lg/"1| values were then plotted on time versus concentration curves by RSTRIPR. Bronchial secretion samples required adjustment for fluid volume differences before analysis and valid comparisons could be made. This was done by converting the zone size to an absolute volume of dmg contained on the disc. rather than a concentration, as it was the amount of drug and not the volume of fluid in which it was dissolved that determined the bacterial inhibitory potential of the compound. Using this data the standard curve for each sample day could be adjusted for absolute drug content per disc regardless of fluid volume (see results). The drug contained in each bronchial secretion sample disc was then read from the linear regression plot of drug vs. zone size using a hand held computer“. These new values were then adjusted for plate to plate variation in the same manner as the fluid samples. Day 1 and 4 samples of BRS and TCF at 1.1 mg/kg were not analyzed by bioassay, but by HPLC at The Upjohn Company, Kalamazoo, Michigan, using reported techniques (Jaglan et al., 1990). Adjustment of this HPLC data was required in order to compare the values from these data sets to all others which had been evaluated by bioassay. Regression analysis of data for which both HPLC and bioassay values were available yielded a mathematical statement from which bioassay equivalent values could be calculated for those sets where only HPLC data had been generated. This calculation yielded data consisting of a mean value rather than individual values for each animal. Therefore, no standard error values are available for this data, and the random factor of individual calves could not be incorporated in the statistical model. n“ Model HP 110, Hewlett-Packard, Corvallis, Oregon. Pharmacokinetic Analysis Pharmacokinetic analysis was performed by microcomputer using RSTRIPR software”. The data entered for analysis was the average of the 4 animals for each sample time and type. All data sets were processed as binomial expressions during curve fitting to the 2 compartment open model (Kinabo and McKellar, 1989), without the variable of weighting factors. This T [1181’ analysis generated the pharmacokinetic values T 1,213, O [DIX’ and AUCMZM, and the statistical values of standard deviation, coefficient of determination and correlation coefficient. The statistical values are useful for determination of how well the Least Squares Regression curve fitting approximates the actual data points. Statistical Analysis Four factors influenced the variables of C T [381’ [1181' T1,”, and AUC: dosage (1.1, 2.2, or 4.4 mg/kg); dose number (single vs multiple); sample location (serum vs TCF vs BRS); and individual calves. The random factor of the individual calves was removed from the model by the use of mean values, therefore three-factor analysis of variance (ANOVA) was done by microcomputerpp according to the general linear model (GLM): Yijk = mean+ A1+Bj+ABU+Ck+Acik+BCJk+ABCiJk(Error) Where Y is the individual variable, A, B, and C were the fixed effects of dose number, dosage level, and sample location, and i, j, and k are levels of each of these factors. The interaction term ABC was used as the systematic error term. Ratios of AUCMZM values were analyzed using factors A and B in a two-factor analysis of variance. Differences in data values were considered significant where the probability of type 1 error (p) was less than 0.05. °° Micromath lnc., Salt Lake City, Utah. PP Number Cruncher Statistical System, Kaysvilie, Utah. 57 Where GLM ANOVA indicated significant differences existed, comparisons between specific means was done using Tukey’s T test. Again, significance was set at p < 0.05. Minimum Inhibitory Concentration Determination of the minimum concentration of ceftiofur required to inhibit growth of 207 clinical isolates‘i‘”I was done using a microtitration, broth-dilution technique (Thmpp, 1986). Ninty six well, round bottom microtiter plates“ were prepared with two-fold decreasing concentrations of the drug, from 10 pg/ml to 0.078 pg/ml, in vertical columns 1 through 12. The dilutions were prepared from 100 mls of 1,000 pg/ml solution, each well being charged with 50 pls of the appropriate concentration. All dilutions were made in sterile french square bottles using double distilled water that then passed through a carbon filter canister, an ion exchange canister, and a 0.22 micron filter" The 8th row (bottom row) of the plate was used as the positive growth control, containing no antimicrobial solution. inoculates of all organisms except :1. somnus were prepared by growing overnight on blood agar at 37 C, then transferring two to five colonies to Mueller-Hinton broth (MHB)“' for three to five hours of incubation at 37 C in a shaking water bath“‘1 at 50 rpm. After incubation the broth cultures were adjusted by addition of fresh MHB to a density of approximately 10° colony forming units (CFU) per ml by comparison with a 0.5 McFarland standard. This suspension qq isolates were obtained from: Animal Health Diagnostic Laboratory, Michigan State University; South Dakota State University, Animal Disease Research and Diagnostic Laboratory, Brookings, South Dakota; Texas Veterinary Medical Diagnostic Laboratory, Amarillo, Texas. H No. 25850, Corning Glass Works, Corning, New York. “ Millipore Corporation, Bedford, Massachusetts. "‘ Difco Laboratories, Inc, Detroit, Michigan. ““ Model 224, Fisher Scientific. 58 was further diluted with MHB by a factor of 1:1000, first by addition of 100 pls of the 0.5 McFarland equivalent to 900 pis of fresh broth, a 1:10 dilution, then by addition of 10 pls of the 1:10 dilution to 990 ”is of MHB to produce a 1:100 dilution. The final broth suspension contained approximately 105 CPU per ml. Haemophilus somnus isolates were prepared by incubating overnight in 5% CO2 on PPLOW agar suppiimented with 1% isoVrtaleXT“. Approximately five colonies were then transferred to PPLO broth with 1% isoVitaieXM and incubated for 5 hours in 5% 002. Dilution to 0.5 McFarland equivalent was also done in PPLO broth with 1% lsoVitaleXTM by the same method as with other organisms. Fifty MS of the final dilutions of all organisms were added to the 50 pls of antimicrobial solution already in each well of columns 1-9, resulting in a final concentration in the microtiter plate wells of 10" CFU/ml (1 :2 dilutions). The last well in each row (Column 12) received 50 pls of broth media, serving as the negative growth control. Control organisms _lg. gneumoniae ATCC 10031 and Mimoccus aureus ATCC 25923 were processed in the same manner and included in columns 10 and 11, respectively, on each plate. Klebsiella pneumgniae is a Gram negative organism that is sensitive to most cephalosporins, while this strain of §. aureus is a Gram positive organism that is quite resistant to ceftiofur sodium. These bacterial organisms serve as positive inhibition and negative inhibition controls between plates. All dilutions were made with Rainin semi-automatic pipetters”. The plates were incubated overnight at 37 C under aerobic conditions, and read visually with a magnifying mirror. The MIC was recorded as the lowest concentration in which no visible growth could be observed. W Difco Laboratories, Inc, Detroit, Michigan. "' BBL Microbiology Systems, Cockeysville, Maryland. ’3 Rainin Instrument Co. lnc., Wobum Massachusetts. RESULTS The procedure for adjusting bronchial secretion samples for fluid volume was first performed on the standard curve data Each standard curve sample disc contained 20 pis of a microgram-per-milliliter dilution, therefore the amount of drug per disc was 0.02 times the concentration value. The absolute (abs) amount of drug for each standard curve dilution is given in the table below. Conc.(uqlmi): 5.0 2.5 1.25 0.625 0.3125 Abs. (pg/disc): 0.1 0.05 0.025 0.125 0.0063 The adjusted bronchial secretion concentrations were calculated from these standard curve values, then used to generate the pharmacokinetic data reported in table 1. The statement used to convert HPLC data to bioassay equivalent date was: Y = 0.937x - 0.789 (r2 = 0.85) Where: Y is the bioassay equivalent value 0.937 is the slope of the regression plot x is the HPLC determined value 0.789 is the Y axis intercept of the regression plot 59 25" L25" LO ' .625? b - mnn . 0 JMZSr 26325 8:228:00 l 25 L 20 1 l5 .Ol Zone (mm) kmmflataunidurammutcdmmtrndmn lamunpdmcfudImnflyhdamnanuysauflww Figure4.Seml-iogpludbacterid 61 Ceftiofur sodium was well absorbed from the injection site in the gluteai muscles of the calves as indicated by peak serum concentrations observed within 0.42 hours to 2.48 hours after injection, depending on dosage and treatment day (Tables 2 and 3, Figures 510). The serum Tm“ values differed significantly (p<0.05) when comparing dosages for the same treatment day, but not between day 1 and day 4 for the same dosage level (Table 2, Figure 11). Maximum serum concentrations (Cum) showed significant linear increases with doubling of the dose (p<0.05), (Tables 2 and 3, Figure 11). Distribution into the tissue chamber fluid occurred slowly, with peak concentration occurring at between 7.13 and 11.26 hours, depending on dosage level and treatment day (Tables 2 and 3, Figures 5-10, 12). Significant TCF Tm” differences were not detected between treatment days for the same dosage, but were observed between dosages on the same treatment day (p<0.05) (Tables 2 and 3, Figure 12). Values for C,m were significantly less in the TCF than in the serum at all dosages and treatment days (p<0.05) (Tables 2 and 3, Figures 5-10). Values for Cam in the TCF increased significantly from low dosage to higher dosage (p=0.01) (Tables 2 and 3, Figure 12). Bronchial secretion concentrations of ceftiofur sodium were significantly higher than TCF, but significantly lower than serum concentrations (p<0.05) (Tables 2 and 3, Figures 5-10). Peak concentrations in BRS occurred much earlier (T max) than in TCF, with Tm“ values much nearer the serum values, although significant differences existed between all sample types (p<0.05) (Tables 2 and 3, figures 5-10). As with TCF, Cmax values for BRS samples increased significantly from low to higher dosage (p<0.05) (Tables 2 and 3, Figure 12). Comparisons between the 3 sample types showed significant increases in Cm, from TCF to BRS to Serum (p<0.05), with Tm, values increasing in the opposite order: Serum to BRS to TCF (p<0.05) (T ables 2 and 3, Figures 510). 62 Ceftiofur elimination half-lives (T1 ”3) showed no significant differences for any comparisons between day, dosage, or sample type. Serum elimination half-life values varied from 1.93 to 3.56 hours, resulting in removal of over 90% of the drug by 8 to 16 hours after dosing. Elimination from BRS was slower, and from TCF slower still, resulting in prolonged drug concentrations in these peripheral compartments (Tables 2 and 3). The AUCO_12hr values expressed as a percentage of serum values showed no statistical differences for either peripheral compartment, although penetration of ceftiofur sodium into TCF increased with higher dosages, while penetration into BRS decreased, for both treatment days (T able 2 and Table 3). Minimum inhibitory concentration results (r able 4, Figures 11, 12, 13) indicate that the concentration of Ceftiofur attained in all sample types when administered once daily at 22 or 4.4 mg/kg is well above the inhibitory concentration for all isolates of common bovine respiratory pathogens, regardless of geographic origin. 63 Table 2. Day 1 pharmacokinetic values for serum, tissue chamber fluid, and bronchial secretions (1 standard error of the mean) collected from 4 calves after once daily intramuscular ceftiofur sodium injections. DOSAGE DOSE LEVEL NUMBER (mg/kg) (day) SERUM‘b" TCF“c BRS“c Cums/ml): 1.1 1 33010.75 0.28" 2.26‘l 2.2 1 8.781247 0.981016 3.951081 4.4 1 17.251935 29512.70 7.821238 T...(hrS)" 1.1 1 1.441031 8.62‘ 1.29d 2.2 1 18210.89 10.061137 2.471138 4.4 1 24810.42 11.261538 2.201380 Twain) 1.1 1 1.931031 8.10cl 1.76‘I 2.2 1 33610.70 11.291535 6.281137 4.4 1 33010.89 193612.27 51511.18 AUC.m,,(pg-hr/mi)° 1.1 1 164612.13 4.87d 19.76‘l 2.2 1 66.17121.63 292011138 470415.16 4.4 1 1319414702 111.8716733 78.1213785 AUC(% of serum value) and» ii ii if ii 1.1 1 29.594 120.04‘l 2.2 1 44.13 71.09 4.4 1 84.79 59.21 C,m values differ significantly between all dosages and types (p < 0.05). Tm values differ significantly between all dosages and types (p < 0.05). AUC values differ significantly between day, dosage and types (p < 0.05). Indicated TCF and BRS values derived by regression analysis of HPLC data against bioassay data. Table 3. DOSAGE Day 4 pharmacokinetic values for serum, tissue chamber fluid, and bronchial secretions (1 standard error of the mean) collected from 4 calves after once daily intramuscular ceftiofur sodium injections. LEVEL (mg/kg) Cums/1110' T...(hl‘S)” T%B(hrs) AUC‘mnng.hr/ml)c AUC(% of serum value) 000‘” II II ll II 1.1 2.2 4.4 1.1 2.2 4.4 1.1 2.2 4.4 1.1 2.2 4.4 1.1 2.2 4.4 Cm values differ significantly between all dosages and types (p < 0.05). T,m values differ significantly between all dosages and types (p < 0.05). DOSE NUMBER (day) #:5-5 &&& «515.5 4 4 4 64 SERUM“""° 5.171055 13.131236 240817.45 0.421025 1.721068 23110.68 3.551030 3.101053 3.311099 29.681482 886511604 1783514433 TCF” 0.28d 1.951038 39411.49 5.77d 100011.49 3.211202 2.07(1 14.121231 101012.13 6.48‘1 703112273 1322014457 21.83‘1 79.65 74.12 BRSW 180" 2.9011.37 5.3610.97 239" 3.27_+_ 1.39 23211.18 5.62" 133011.75 48411.17 20.10d 68.181526 554113.34 67.72d 76.91 31.07 AUC values differ significantly between day, dosage and types (p < 0.05). Indicated TCF and BRS values derived by regression analysis of HPLC data against bioassay data. Concentration lug/ml) 100 W i 1 1O 1 II fill" r I IIIHII 0 1 a s 4 s s 1 a s 101112 Timefhours) * Serum -+- TCF + BRS msmwmmmmm.wmumrens) cortcerlrationsofeefiiofixsodiunh4ealvesdter1dosen1.1mglkg.*- wmmmaflonwdmdmdmfid. Concentration (tug/ml) 100: i- i" 1.. l- ‘OEE. +- H—‘JQ c ‘- 1 + : 1 p- r 01 j l 1 I l l I J I I C 012 s 4 s 0 1 a s 10 1112 Time(h0urs) —‘- Serum + TCF + BRS Flgme806ul¢edsumflssued1ambaflddfl0flmdbmndiflseasfloan$ concentrationscfeeftiofusodiunh4ealvesdter1dosel22mglkg. *- WMWWpJ-umamng. Concentration (no/mi) 100: I +- p- 105 . h- _c : . - ‘_L + ‘ ‘-‘) 15 :. s. h P P 01 iiJlljllLLJ O 012 s 4 e e 7 e s 1011 12 Time (hours) —°" 80mm + TCF ‘9‘ IRS Figwe7.CdalaedaenntflssuedtambuMd(fCFi.mdbrmssaefion(BRS) Wdcfllofuaodunh4esivesdter1dossls4num. *- wmmmmgmnnodmnd. Concentration (fig/ml) 100 A . 0.01 — l ‘ ' 1 ' 4 I I 1 1 1 O 1 2 3 4 5 G 7 B 9 10 11 12 Time (hours) * Serum ‘4— TCF “'9- BRS Flgure8.Calculatedsen1m. tissue charnberlluid (TCF), and bronchial secretion (BRS) concentrationsofceitiofursodiwnh4eslvesdter4dosesnt1mg/kg '- signlicarniydflerernmadmmneuMatim(C_)mdtRMdoecum(r_). Concentration (us/ml) 100: r- 10;, .. 01 L J I l I I l l J l I O 1 2 3 4 5 O 7 8 O 10 11 12 Time (hours) * Serum -+- TCF + BRS Figuree.Cdulatedsermtflssued1arnbeer(TCF).mdbrmcflalseueflon (BRS) eorreemationsofeeftiofusodiumhscmesafter4dosesu22mglkg. *- slgnlicemudfieremmaadmuncmcematimwdandfimedoecwancefld. Concentration (no/ml) 100 1 1111”] 1H)” — _ h h i- i— b r- i- °0 1 2 s 4 s s 7 s s 10 11 12 Time(hours) -°— Serum -+— TCF + BRS Figum10.Caiqfledsamtflssued1amberfluldUCF).mdmaeaefion(BRS) concertsflonsofceftiofusodiumh4eeivesafter4d0sesl¢4mglkg *- dgmaimmconeemmioMC-imdmdmCLJ. n 100 Cones tration (ug/mi) 71 ‘7 ....................................... ....... ........... 3 ............................... a ...... a b .. ................................................................................................... 0 .................................................................................................... d 0.1 ................................................................................................... 0 0.01 ' ‘ 1 ' ‘ ' T 0 2 4 0 0 1O 12 14 Figure 11. Legend: Time (hours) ‘9“ 1.1 Ila/kn ‘9‘ 2.2 MM ‘9' 4.4 lilo/kg Calculated serum concentrations of ceftiofur for 4 calves after 4 doses. compared to minimum inhibitory concentration values for 90% of clinical bacterial isolates (MiCm). 8 " Lnatia ____marcescens, W m b s [ggbsielia gggq Escherichia my, Entggbacter £29. 0 - W snaffle d - Stregococcus mggalactia 9 " £____asteurella We") 1 W m 11W m * - Significantly different maximum concentration (C_n) and time of occurance 1r...)- 10 Concentration (no/ml) 0.0 1 2 4 s s 10 12 14 Time (hours) +11M0M +£2.01” +4-4MM Figure 12 CdcuatedTCFeamraionsdceffidmfor4ceNesdter4doses,comparedto Legend: minimum hhibitory concentration values for 90% 0f cinical bacterial isolates (MIC...)- a- $1202me b- metamsmemmm 3' WW2 I l' - fig gggllg mgmsmca' EM M2002. 8mm: ma :- Signzicamydfliererumaadnamcmcemaionmgmdmdmance 0.0. 10 0.1 Concentration (no/ml) r T 1111” 0.0 1 j l I l l l 0 2 4 s s 10 12 14 Time (hours) +1.1mgllra +2.2mgllra +4.4 rug/Ira Figure 13. Calculated BRS concentrations of ceftiofur for 4 calves after 4 doses, compared to Legend: minimum inhibitory concentration values for 90% of clinical bacterial isolates (Mica) ' " Lmiam.mm 9' ISLQL"e ienaMquWM- c- WM d- WW 0- mmmmmm '- Signflicamydifierersmaadmuncmcemation(c_)mdfimecfoccurance (T...)- Table 4. 74 Cumulative and (percentage of total) in vitro minimum inhibitory concentrations of ceftiofur sodium for bacterial isolates recovered from cattle. ORGANISM Pasteurella haemolygica Pasteurella multocida Eschericha go_ii Haemophilus somnus Staphylococcus aureus Actinomvces pyogenes Streptococcus uberis Klebsiella s29. Streptococcus agalactia Serratia marcescens Enterobacter cloacae Enterobacter fecalis Listeria monocytogenes NUMBER TESTED 64 42 .2de rororororocioaogmo <0.078 64(100) 42(100) 25(1 00) 10(100) 6(100) 1 (50) Ceftiofur concentration (pg/ml) 0.158 0.3125 0.625 1.25 8(27) 24(80) 27(90) 30(100) 2(14) 1 1 (79) 14(100) 2(33) 6(100) 2(100) 1(50) 2(100) 1(50) 2(100) 2 2.5 DISCUSSION This study applied tested and accepted methods in two novel ways. One was in the pharmacokinetic evaluation of a new antimicrobial drug, the other was in evaluation of a new respiratory secretion sampling technique. Pharmacokinetic evaluation of Ceftiofur sodium 0' ables 2 and 3) indicates that the dmg is well absorbed and reaches appreciable concentrations in systemic circulation after intramuscular injection. The values measured in this study also compare closely with those generated in the research and development of this compound for the same sample types and dosage levels, with the exception of respiratory samples. This difference will be discussed further below. Although no reports of pharmacokinetic analysis of this drug exist, other third generation cephalosporins have been studied in animal and human subjects (Faulkner et al., 1987; Nakashima et al., 1989; Soback and Ziv, 1988; Soback and Ziv, 1989; Hoffstedt and Walder, 1981). Most studies with human subjects involve oral or intravenous administration of antimicrobial agents, with pharmacokinetic measurement of the central compartment only. Oral and M pharmacokinetic data differ from IV data in that no lag time exists when drugs are administered by N routes. Bolus intravenous administration also results in the entire dose of compound being placed directly and immediately into the systemic circulation (Prescott and Baggot, 1988a). in contrast, oral and iM administration result in delays before measurable concentrations of drug appear in serum (Greenblatt and Shader, 1985; Prescott and Baggot, 19883). Orally administered drugs may also be subject to variations in stability to gastrointestinal contents such as protozoa, bacteria, and enzymes, and are variably absorbed 75 76 imo the system dependant primarily on the lipid solubility of the drug (Prescott and Baggot, 1988a). These factors may result in substantially poorer drug absorbtion when administered by oral compared to IM routes. Orally administered compounds are also subject to metabolism by gut mucosa or liver before reaching the systemic circulation, again, influencing the amount of active compound present in the systemic circulation (Prescott and Baggot, 1988a). Intramuscular admininstration, however, does not assure uniform absorbtion. Vascular perfusion and capillary blood flow, the primary determinants of absorbtion, vary between muscle groups, resulting in different drug absorbtion rates (Prescott and Baggot, 1988a). These factors affect the pharmacokinetics of the drug, making comparisons between drugs difficult or meaningless if the technique and route of administration are not identical. For these reasons, comparison of ceftiofur sodium data from this study to data from human studies of other cephalosporins cannot be made. However, most investigations of cephalosporins in animals involve similar techniques and therefore yield useful commrisons. in studies of cephalosporins involving animals, ceftriaxone has been shown to reach peak serum concentrations of 25 149/1111 and 35 pg/ml within 30 minutes of a single intramuscular dose to calves at 10 mg/kg and 20 mg/kg, respectively (Soback and Ziv, 1988). Cefoperazone reached a concentration of 16 pg/mi of calf serum within 30 minutes of single dose intramuscular administration at 20 mg/kg to unweaned calves (Soback and Ziv, 1989). The dosages in these investigations exceeded the maximum dosage of ceftiofur in our study, which is reflected by larger Cm, values. Serum Cm, values for ceftriaxone and cefoperazone occured earlier than our semm Cm values (indicated by lower Tmax values), suggesting that ceftiofur is absorbed more slowly than these other third generation agents. Elimination half-life values after single doses of ceftriaxone at 10 mg/kg and 20 mg/kg, and cefoperazone at 20 mg/kg were 116.8 minutes, 145 minutes, and 136.9 minutes, respectively. Semm T1/zp values for ceftiofur ranged from 115.8 minutes after a single dose of 1.1 mg/kg, to 213.6 minutes following a single dose of 2.2 mg/kg. These differences may be the result of variations in protein 77 binding. Ceftiofur is reported by the manufacturer to be approximately 90% bound in bovine serum. Ceftriaxone was estimated to be only 22% bound at the low dose and 18% bound at the high dose (Soback and Ziv, 1988). Increased serum protein binding delays diffusion from the vascular space to the tissue spaces (Craig and Suh, 1985. Highly protein bound drug that does reach the tissue space is also readily bound by tissue proteins (Craig and Suh, 1985). The result of these interactions is reduced tissue peak concentrations and delayed clearance (Craig and Suh, 1985), as seen in these investigations. These observations on the effect of protein binding are supported by conclusions following the examination of cephalosporin penetration into human subcutaneous tissue fluid, using a subcutaneous cotton thread implant model (Hoffstedt and Walder, 1981). This study involved 5 cephalosporins with serum protein binding values ranging from 10% to 90%, and demonstrated that the more highly protein bound drug also had the longest elimination half- life, both in serum and interstitial fluid, irrespective of whether the cephalosporin was administered by intravenous bolus or infusion. This study showed that while ceftiofur penetrates into TCF at concentrations that seem adequate for inhibition of susceptible bacteria that might invade this environment, it does so slowly and at concentrations substantially lower than those in serum, as indicated by the longer Tu,“ and smaller Cum and AUC values. This is consistent with the results reported by Hoffstedt and Walder where peak tissue fluid/peak serum levels were 41% for low protein bound (10%) cephradine, compared to 5% for highly serum protein-bound (90%) cefoperazone. Even with a relatively low serum protein binding value, cephradine (a first generation cephalosporin) was only able to achieve 41% of the serum concentration in the extravascular fluid. While it has been consistently observed (Chisholm et al., 1973; Joiner et al., 1981; Short et al., 1987) that the extravascular fluid evaluated in tissue chamber studies has lower peak 78 concentrations that occur simultaneously with or later than serum peaks, and that such fluids have longer T1,” values consistent with those generated in this investigation, this phenomenon has not been completely explained. Possible explainations may involve tissue protein binding, molecular size, pH, or ionic restrictions to diffusion. The tissue chamber model has met with acceptance as a method of tissue fluid collection, as is indicated by the number of published reports using this system in various related forms (Clarke et al., 1989a; Peterson et al., 1989; Higgins and see, 1984; Short CR et al., 1987; Walker et al., 1989; Walker et al., 1990). Although fluid collected from these chambers appears to be quite similar to natural, non-inflamed interstitial fluid (Chisholm et al., 1973), it is collected as a result of an artificial tissue space and may be influenced by the presence of the device or the space (Bergen, 1981). Factors other than protein binding may also influence tissue penetration of antimicrobial agents. Tissue fluid pH, for example, has been shown to effect antimicrobial concentration in tissue chambers implanted in the prostate glands of dogs (Bergan, 1981). Canine prostatic fluid is acidic; acidic compounds such as penicillins and cephalosporins penetrated less well into prostatic fluid than into the subcutaneous tissue chamber fluid, while the alkaline compounds erythromycin and trimethoprim produced concentrations that exceeded tissue fluid levels (Bergan, 1981). These differences suggest that specialized tissues such as glandular or secretory tissues may have properties that require separate consideration from such generalized tissues as skeletal muscle or subcutaneous interstitial tissue. Even though subcutaneous tissue is considered a generalized tissue, precise characterization of TCF for pH, lipid, protein and ion content, coupled with investigation of the behavior of the assayed compound under similar conditions, would allow the most accurate interpretation of the data (Clarke et al., 1989). These factors were not considered in the current study, and may have played a role in the observed low TCF ceftiofur sodium concentration, delayed Tm“, and long T1,”. 79 Another factor which may influence these parameters is the perforated surface area of the tissue chambers. The chambers described by Chisholm et al. (1973) had approximately 40% of the surface area opened by perforations, those used by Walker et al. (1989, 1990) were approximately 40% open. The chambers in this study had approximately 20% perforated surface area. Since the surface area of a cylinder varies as the square root of the volume, larger chambers will have smaller areazvolume ratios relative to small chambers, yet the larger cylinder will have greater volume. if these chambers function as reservoirs (Bergan, 1981), delays in equilibrium will occur between chamber lumen contents and the surrounding fluid or interstitial gel, with larger chambers or smaller perforated surface area values resulting in longer delays. While these chambers are felt to be in direct communication with interstitial space, the distances between capillaries, lymphatics, interstitial tissue, and interstitial spaces; the dynamic components that are responsible for the composition of interstitial fluid, are minute compared to the distances between those components and the tissue chamber lumen. This delay in equilibrium may be responsible for the prolonged tissue chamber Tm: and T1 ,2, seen in this study (T ables 2 and 3, Figures 5-10, 12). Bioassay-equivalent values for the low dosage (1.1 mgkg) were generated mathematically from HPLC data and may be subject to some question in comparison to the actual bioassay data generated for other dosages, although the importance of this information makes these comparisons necessary. The HPLC analysis was conducted on pooled samples, therefore the effect of one animal's sample exerting an unusually large effect on the total measurement cannot be evaluated. The influence of a single sample on the pharmacokinetic values for the pooled sample measurement may have been responsible for some inconsistency in the values for this data set. Ceftiofur sodium is marketed as an antimicrobial agent for the treatment of bacteriaily-associated respiratory disease in cattle. The importance of understanding the pharmacokinetics of this 80 drug in the infected tissues has been discussed. Our results show that ceftiofur sodium is capable of attaining high concentrations in the bronchial secretions soon after administration. The values shown in this study appear higher than the those measured in the ceftiofur sodium researach and development, but may be explained by the differences in experimental methods. Peak BRS concentration of 3.95 pQ/ITII occured at 2.47 hrs for the 2.2 mg/Ib dosage (T able 2) in this study. Lung tissues sampled at 8 hours only in the development of this compound yielded concentrations of 0.9 - 1.4 pg/ml. Examination of Figure 6 shows BRS concentrations declining to near this value by 8 hours, with remaining differences falling within the error range reported in this study. Differences in sampling techniques (BRS swab vs. tissue plugs) make further comparisons difficult. Bronchial secretion Cw: values were significantly higher than the corresponding TCF samples, and showed significant increases from low dosage to higher dosage. interestingly, values for sample day 1 exceeded day 4 values for all dosage levels, although the differences were not statistically significant. The bronchial secretion sampling technique, like the tissue chamber technique, may influence the composition of the sample and perturb the results by some unexplained mechanism. Alternative sample methods, however, are likely to influence the outcome as well. Biopsy and homogenation techniques have been discussed, and found to be unsatisfactory. Lavage techniques (Gray et al., 1989) require compensatory fluid dilution adjustments, and probably alter subsequent samples by dilution and inflammation similar to the disc absorption method. Pulmonary mechanical abnormalities have been observed following small volume (20 ml) fluid instillation into tracheas of calves (Killingsworth et al., 1987). This effect is felt to be at least partially mediated by the vagus nerve via irritant and chemoreceptors; other parasympathetic-mediated effectors in the lung include pulmonary and vascular smooth muscle, secretory cells, and inflammatory mediators (West, 1982). These autonomic responses have the potential to significantly modify the bronchial secretions (Murray, 1986). 81 Other respiratory secretion swab techniques involving rigid structure contact with bronchial mucosa would be just as likely to induce changes in the secretions by these same mechanisms (Hajer et al., 1988). Evaluation of other pulmonary fluids (pulmonary lymph) would not necessarily reflect drug activity at the airway surface, where _P_. haemoMica attacks, but would possibly provide useful information regarding pulmonary parenchymal drug concentrations, and might be collected by microprobe cannuiation that would have minimal influence on sample composition. Differences in lymph, serum, and interstitial fluid have also been discussed. As stated previously, successful antimicrobial therapy requires that concentrations of non-protein bound dmg exceed the MIC of the pathogen at the site of infection for (Prescott and Baggot, 1988b; Sande and Mandell, 1985; Wise, 1985). The antimicrobial agent should be administered such that the free drug concentration at the site of infection exceeds the MIC target values for the entire treatment period, since inadequate (subinhibitory) beta-lactam antimicrobial levels have been associated with 'breakthrough' of susceptible bacterial strains during the first 72 hours of therapy (Anderson et al., 1976). inadequate dosing may also provide evolutionary pressure towards development of, or selection for resistance in previously susceptible bacterial strains (Bellido et al., 1989). The resistance that develops in this manner may not be specific for the antimicrobial agent in use, but may be more broad in spectrum based on binding site or membrane permeability modifications (Bellido et al., 1989). The risk of 'breakthrough’ resistance development is greater with antimicrobials that do not induce a significant post- antibiotic effect (PAE), which is defined as the persisting suppression of bacterial growth that follows limited exposure to an antimicrobial agent. Post-antibiotic effects vary dependant on specific drug, organism, and the amount the Cmam for that drug exceeds the MIC for that organism (Vogeiman and Craig, 1985; Craig and Gudmundsson, 1985). Beta-Iactam antimicrobials are not considered to induce significant PAE against gram-negative organisms (Vogeiman and Craig, 1985; Craig and Gudmundsson, 1985). These drugs require more careful 82 monitoring and observation of dosing guidelines as determined by pharmacokinetic behavior than do those inducing post-antibiotic effects. Maintaining biologically active, non-protein bound drug concentrations above the MIC for the target organism throughout the dosing period has been shown to be the most important parameter in determining beta-lactam therapeutic efficacy (Vogeiman et al., 1988). Because MIC determination is done in vitro, the actual tissue drug concentration required to inhibit the organism is unknown, being affected by pH, protein binding, ionic binding, penetration restraints, and other influences (Clarke et al., 1989), and varies with the dynamic processes occurring at the target location. Therefore, drug administeration dosages and intervals should maintain concentrations of at least four times the MIC90 for the pathogen throughout the treatment period (Sande and Mandell, 1985). Care must be taken with some dmgs, such as aminoglycosides, to ensure that this does not place the drug concentration in the range of toxicity. Fortunately, beta-lactam drugs have a wide therapeutic margin (Balant, 1985), and this is rarely the problem. Although toxicity was not a subject of this investigation, existing dataW indicates that a wide margin of safety exists between therapeutic and toxic concentrations with ceftiofur sodium. Our data suggests that ceftiofur would be an appropriate choice as treatment for soft tissue, respiratory or bacteremic infections in calves caused by E. haemoMica, E. multocida, A gyggenes, and _S_. gbefls. All samples, (serum, TCF, and BRS) and dosages (1.1, 22, and 4.4 mg/kg) yielded drug concentrations at least 4 times greater than the MIC90 of 0.078 jug/ml generated in this study for these bacterial agents 0' ables 2, 3, and 4). The MIC data presented in this study corresponds well with the values generated in the research and development of this drug (I’ able 1), and are supported by a current report (Yancey et al., 1987) in which the ’7 Miller, C.C. Ceftiofur research and development. Personal communication. 83 MIC values for ceftiofur sodium and ampicillin were compared, finding that the MIC values for ceftiofur were lower than the ampicillin values. Ceftiofur was also more active than ampicillin or cefamandole, a second generation cephalosporin, against a multiple-antibiotic resistant strain of _P_. haemoMica (Yancey et al., 1987). The tissue concentrations achieved by once daily dosing in this investigation also persisted to 12 hours after dosing, satisfying the recommendations of maintaining drug levels of four times the MIC90 value throughout the dosing period (Figures 11, 12, 13). Because of the overall broad spectrum of the cephalosporins, treatment of bacterial infection caused by pathogens other than those commonly associated with BRD might be attempted. These applications would require close evaluation and monitoring. For example, ceftiofur as treatment of _S_. eggs-induced cellulitis could not be recommended based on this study and stated guidelines since the MIC90 value generated in this study for this organism is 1.25 pg/ml. This value is close to the 2.2 mg/kg., day 4 peak value of 1.95 jug/ml, and although this MIC90 value is more than twice exceeded by the peak values generated by the 4.4 mg/kg dosage regimen, it does not provide the therapeutic ratio that Is considered acceptable, and other drugs with more consistent and lower MIC90 values would be better choices. This same logic would apply for infections caused by Klebsiella s99, most g. M strains, and m marscesens, although the number of Klebsiella and Serratia isolates evaluated in this study were limited, and should be considered as general indicators only. Other organisms such as _LiiegLa monflogenes and Enterobacter fe_caLs appear resistant to the ceftiofur sodium, although, again, very few isolates were evaluated. Specific bacterial culture and antimicrobial susceptibility evaluation of these organisms would indicate alternate, and more appropriate, antimicrobial choices. Choice of specific antimicrobial agent, or scholastic comparison of similar compounds, might be facilitated by studies of serum bactericidal activity over time rather than estimating efficacy based on half-life, peak concentration and in vitro bacterial inhibition data as is the current standard (Gugleilmo and Rodondi, 1988). 84 The data generated in this study indicates that ceftiofur sodium is an appropriate antimicrobial agent against the bacteria associated with BRD. Increasing the dose from 1.1 mg/kg to 22 or 4.4 mg/kg, or dosing at 12 hour rather than 24 hour intervals, however, would provide larger therapeutic ratios, maintaining the drug concentration above the MIC90 value throughout the dosing interval in all tissues examined. As indicated by T1,”, tissue chamber concentrations and bronchial secretion concentrations of ceftiofur sodium will still exceed MIC levels by recommended margins after 24 hours, when closed at 2.2 or 4.4 mg/kg, although serum concentrations would decrease to sub-inhibitory levels in this amount of time. As stated earfier, exceeding the MIC value by greater margins is likely to provide no therapeutic benefit (for drugs that induce little or no PAE) other than prolonging the length of time that drug concentrations exceed this value, again, ensuring that concentrations exceed MIC values during the entire dosing period (Vogeiman et al., 1988). Lower dosages given more frequeme would be as effective, and while toxicity does not appear to be a concern with this drug, reduced dosage therapy might be important with other antimicrobial agents (Vogeiman et al., 1988). The manufacturer’s recommended dosing interval of 24 hours for ceftiofur is, incidentally, an exception compared to the majority of the third generation cephalosporins. Intervals of 12 hours, with shorter intervals of 4-8 hours in life threatening situations, are recommended for most of the other cephalosporins (Belem et al., 1985). Therefore, treatment with 2.2 or 4.4 mg/kg at 24 hour intervals appears acceptable when the target organism is known and is susceptible at low drug concentrations, and when drug penetration into tissues can be expected to be consistent with the healthy subjects of this study. In debilitated or chronically diseased animals drug penetration would be expected to be reduced due to dehydration, circulatory debilitation, and chronic inflammatory change (fibrosis), in which case higher dosages or more frequent dosing intervals might be necessary for satisfactory results. This would also be true for more resistant organisms, where culture and susceptibility date would be indicated for most accurate therapeutic approach. LIST OF REFERENCES LIST OF REFERENCES Al-Derraji AM, Cutlip RC, Lehmkuhl HD, Graham DL Experimental Infection of lambs with bovine respiratory syncytial virus and Pasteurella haemolflica: Pathologic studies. Am J Vet Res 42: 224-229. Amstutz HE, Horstmen LA, Morter RI. Clinical evaluation of the efficacy of Hemoghilus somnus and Pasteurella spp. bacterins. Bov Pract 1981: 16:106-108. Allen EM, Gibbs HA, Wiseman A, Selman IE. Sequential lesions of experimental bovine pneumonic pasteurellosis. Vet Rec 1985; 117:438-442. Allen EM, Gibbs HA, Wiseman A, Selman IE. Experimental production of bovine pneumonic pasteurellosis. Res Vet Sci 1984; 37:1544 66. Anderson ET, Young LS, Hewitt WL Simultaneous antibiotic levels in 'breakthrough' gram- negative rod bacteremia. Am J Med 1976; 61 :493-497. Babiuk, LA, Lewman MJP, Gifford GA Use of recombinant bovine alpha1 interferon in reducing respiratory disease induced by bovine herpes virus type 1. Antimicrob Agents Chemother 1987; 31:752-757. Babiuk LA, Lewman MJP, Gifford GA A seminar in bovine immunology. Trenton, New Jersey: Veterinary Learning Systems, Inc., 1987; 12-23. Balant L, Dayer P, Auckenthaler R. Clinical Pharmacokinetics of the Third Generation Cephalosporins. Clinical Phermecokinet 1985; 10:101-143. Baluyut CS, Simonson RR, Bemrick WJ, Maheswaran Sk. Interaction of Pasteurella haemoMica with bovine neutrophils: Identification and partial characterization of a cytotoxin. Am J Vet Res 1981; 42:1920-1926. Barry AL Procedures for testing antimicrobials in agar media: theoretical considerations. In: Antibiotics in Laboratog Medicine, 2nd ed. Lorian V, ed. Baltimore: Williams and Wilkins 1986; 1-26. Bellido F, Viadoianu IR, Auckenthaler R, Suter S, Wacker P, Then RL, Pechere JC. Permeability and penicillin-binding protein alterations in Salmonella muenchen: Stepwise resistance acquired during beta-lectam therapy. Antimicrob Agents Chemother 1989; 33:1113-1115. Bergen T. Pharmacokinetics of tissue penetration of antibiotics. Rev Inf Dis 1981; 3:45-66. Bergen M, Weinstein L Penetration of antibiotics into fibrin clots in vivo. J Inf Dis 1974; 129:59- 78. 86 87 Berggren KA, Baluyut CS, Simonson RR, Bemrick WJ, Maheswaran SK. Cytotoxic effect of Pasteurella haemoMica on bovine neutrophils. Am J Vet Res 1981; 42:1383-1388. Bieiefeldt OH, Lawman MJ, Babiuk LA. Bovine interferon: it's biology and application in veterinary medicine. Antivir Res 18=987; 7:187-210. Biberstein EL, lGrkham C. Antimicrobial susceptibility patterns of the A and T types of Pasteurella haemoMica. Res in Vet Sci 1979; 26:324-328. Blood DC, Radostits OM, Hendersen JA. Veterinary medicine: A textLbOOk of the diseases of cattle, sheep, gigsI goats, and horses, 5th ed. London: Dailliere Tindall 1983; 329-330. Breeze R. Stmcture, function, and metabolism in the lung Vet Clin N Am (food en prect), bovine respiratory disease edition. Philadelphia: W.B. Saunders Co. 1985; 219-235. Breider MA, Walker RD, Hopkins FM, Schultz TW and Bowersock TL Pulmonary lesions induced by Easteurella haemolvtica In neutrophil sufficient and neutrophil deficient calves. Can J Vet Res 1988; 52:205-209. Bryson DG. Calf pneumonia. Vet Clin N Am (food an prect), bovine respiratory disease edition. Philadelphia: W.B. Saunders Co. 1985; 237-257. Bryson DG, McNuIty MS, McCracken RM, Ultrestructural features of experimental pereinfluenza type 3 virus pneumonia in calves. J Comp Path 1983a; 93:397-414. Bryson DG, McNuIty MS, Logan EF, Respiratory syncytial virus pneumonia in young calves: Clinical and pathologic features. Am J Vet Res 1983b; 44:1648-1655. Bush K. Characterization of B-lactamases. Antimicrob Agents Chemother 1989a; 33:259-263. Bush K. Classification of B-iactamases: Groups 1, 2a, 2b, and 2b'. Antimicrob Agents Chemother 1989b; 33:264-270. Bush K Classification of B-lectamases: Groups 20, 2d, 2e, 3, and 4. Antimicrob Agents Chemother 1989c; 33:271-276. Chang Y, Renshew HW, Young R. Pneumonic pasteurellosis: Examination of typeable and untypeable Pasteurella haemoMice strains for leukotoxin production, plasmid content and antimicrobial susceptibility. Am J Vet Res 1987; 48:3. 378-384. Chisholm GD, Watersworth PM, Calnan JS, Gerrod LP. Concentration of antibacterial agents in interstitial tissue fluid. Br Med J 1973; 1:569-573. Clarke CR, Short CR, Corstvet RE, Nobles D. Effect of Pasteurella haemoMica Infection on the distribution of suifediezine and trimethoprim Imo tissue chambers implanted subcutaneously in cattle. Am J Vet Res 1989a; 50:1551-1556. Clarke CR, Short CR, Usenik EA. Subcutaneousiy implanted tissue chambers: A pathophysiological study. Res Vet Sci 1989b; 47:195-202. Clarke, SW. Physical defenses of the respiratory tract. Eur J Resp Dis 1983 (suppl); 64:27- 30. 88 Confer AW, Panciera RJ, Mosier DA. Bovine pneumonic pasteurellosis: immunity to Pasteurella haemoyfica J Am Vet Med Assoc 1988; 193:1308-1316. Corbeil LB, Gogolewski RP. Mechanisms of bacterial injury. Vet Clin N Am (food an prect), bovine respiratory disease edition. Philadelphia: W.B. Saunders Co. 1985; 367-376. Corstvet RE, Panciera RJ, Rinker HB, Starks BL, Howard BS. Survey of tracheas of feedlot cattle for Haemophilus somnus and other selected bacteria. J Am Vet Med Assoc 1973; 163:870-873. Craig WA, Suh B. Protein binding and the antimicrobial effects: Methods for the determination of protein binding. In: Antibiotics1in laboratory medicine, 2nd ed. Lorian V, ed. Baltimore: Williams and Wilkins 1985; pg. 477-514. Craig WA, Gudmundsson S. The postantibiotic effect. In: Antibiotics in laboratory medicine, 2nd ed. Lorian V, ed. Baltimore: Williams and Wilkins 1985; pg. 515-538. Curtis NAC. Session 2: Biochemistry of beta-lactamases and penicillin binding proteins. Rev Inf Dis 1988; 712-743. Dellmen HD, Brown EM. Textbook of veterina_rv histology. Philadelphia: Lee and Febiger 1976; 202. Drusano GL Role of pharmacokinetics in the outcome of infections. Antimicrob Agents Chemother 1988; 32:289-297. Duffee NE, Christensen JM, Craig AM. The pharmacokinetics of cefadroxil in the foal. J Vet Pharmacol Ther 1989; 12:322-326. Dyer RM. The bovine respiratory disease complex: A complex interaction of host, environmental, and infectious factors. Compend Cont Ed Pract Vet 1982; 4:S296-6304. Edberg SC. The measurement of antibiotics in human body fluids: techniques and significance. in: Antibiotics in laboratory medicine, 2nd ed. Lorian V, ed. Baltimore: Williams and Wilkins 1986. Emau P, Girl SN, Bmss ML, Zia S. Ibuprofen prevents Pasteurella haemolygica endotoxin- induced changes in plasma prostenoids and serotonin, and fever in sheep. J Vet Pharmacol Ther 1985; 8:352-361. Fales WH, Selby LA, Webber JJ, Hoffman LJ, Kintner LD, Nelson SL, Miller RB, Thome JG, McGinity JT, Smith DK. Antimicrobial resistance among Pasteurella spp. recovered from Missouri and Iowa cattle with bovine respiratory disease complex. J Am Vet Med Assoc 1982; 181 :477-479. Faulkner RD, Bohaychuk W, Desjardlns RE, Look ZM, Haynes JD, Weiss Al, Silber BM. Pharmacokineties of ceflxime after once-a-dey end twice-a-day dosing to steady state. J Clin Pharmacol 1987; 27:807-812. Filion LG, Wilson PJ, Bieiefeldt-Ohmann H. The possible role of stress in the induction of pneumonic pasteurellosis. Can J Comp Med 1984;48:268-274. 89 Frank GH, Nelson SL, Briggs RE. Infection of the middle nasal meatus of calves with Pasteurella haemoMica serotype 1. Am J Vet Res 1989; 50:1297-1301. Frank GH, Smith PC. Prevalence of Pasteurella haemolygica in transported calves. Am J Vet Res 1983; 44:981-985. Frank GH, Briggs RE, Gillette KG. Colonization of the nasal passages of calves with Pasteurella haemoMica Serotype 1 and regeneration of colonization after experimentally induced viral infection of the respiratory tract. Am J Vet Res 1986; 47:1704-1707. Friend SC, Thomson RG, Wilkie BN. Pulmonary lesions induced by Pasteurella haemoMica in cattle. Can J Comp Med 1977; 41 :219-223. Gray PR, Derksen FJ, Robinson NE, Carpenter-Deyo LJ, Johnson HJ, Roth RA. The role of cyclooxygenase products in the acute airway obstruction and airway hyperreectivity of ponies with heaves. Am Rev Respir Dis 1989; 140:154-160. Grey CL, Thomson RG. Pasteurella heemoMica in the tracheal air of calves. Can J Comp Med 1971; 35:121-128. Greenblatt DJ, Shader RI. Pharmacokinetics in clinical practice. Philadelphia: WB Saunders Co. 1985. Gugleilmo BJ, Rodondi LC. Comparison of antibiotic activities by using semm bactericidal activity over time. Antimicrob Agents Chemother 1988; 32:1511-1514. Gutmann L. Ferre B, Goldsteln FW, Rizk N, PInto-Schuster E, Acar JF, Collatz E. SHV-5, a novel SHV-type B-Iactamase that hydrolyses broad-spectrum cephalosporins and monobectams. Antimicrob Agents Chemother 1989; 33:951-956. Hajer R, Wensing Th, Keg R, Boenna S, Willems Fl'hC. An improved technique for sampling bronchial secretions from normal calves. J Vet Med 1988; 35:739-743. Higgins AJ, Lees P. Tissue cage model for the collection of nflemmatory exudate in ponys. Res Vet Sci 1984; 36:284-289. Hjerpe CA. Treatment regimes for feedlot cattle with bronchopneumonia and fibrinous pneumonia. In: Proceedings of 10th annual food animal medicine conference - The use of dmgs in food animal medicine. Powers, J.D., Powers, T.E, eds. Columbus, Ohio: Ohio State University Press 1984; 228-255. Hjerpe CA. Clinical management of respiratory disease in feedlot cattle. Vet Clin N Am (food an prect), bovine respiratory disease edition. Philadelphia: W.B. Saunders Co. 1983; 5:120- 142. Hoffstedt B, Welder M. Influence of serum protein binding and mode of administration on penetration of five cephalosporins into subcutaneous tissue fluid in humans. Antimicrob Agents Chemother 1981; 20:783-786. Hurvitz Al. immunology. In: Textbook of veterina_rv internal medicine. Ettinger SJ, ed. Philadelphia: WB Saunders Co. 1975; 1701-1712. 90 Jaglan PS, Cox BL, Amoid TS, Kubicek MF, Stuart DL, Gilbertson TJ. Liquid chromatographic determination of desfuroylceftiofur metabolite of ceftiofur as residue in cattle plasma. J Assoc Off Anal Chem 1990; 73:26-30. Jensen R, Mackey DR. Diseases 91 feedlot cattle, 3rd ed. Philadelphia: Lea and Febiger 1979. Joiner KA, Lowe BR, Dzink JI., Bartlett JG. Antibiotic levels in infected and sterile abscesses in mice. J Inf Dis 1981; 143:487-494. Jones RN, Barry AL, Bevin TL, Washington JA II. Susceptibility tests: Microdilution and macrodilution broth procedures. In: Manual of clinical microbiology. 4th ed, Lennette EH, Bellows A, Heusler WJ et al., eds. Washington: Am Soc Microbioi 1985; 972-977. Jubb KVF, Kennedy PC, Palmer N. Pathology of domestic animals, 3rd ed, Vol 2. Orlando: Harcourt Brace Jovanovich 1985; 453. Kent JE, Ewbenk R. The effect of road transport on the blood constituents and behaviour of calves. ll. One to three months old. Br Vet J 1986a; 142:131-140. Kent JE, Ewbenk R. The effect of road transport on the blood constituents and behaviour of calves. ill. Three months old. Br Vet J 1986b; 142:326-335. Kerremans AL, Lipsky JJ, Van Loon J, Gailego MO, Weinshilboum RM. Cephalosporin-induced hypoprothrombinemia: Possible role for thiol methylation of 1-methyltetrazole-5-thiol end 2- methyI-I,3,4-thiadiazoIe-5-thiol. J Pharmacol Exp Therapeutics 1985; 235:382388. lfiiiingsworth CR, Slocombe RF, Alnoor SA, Robinson NE, Derksen FJ. Pulmonary dysfunction in neonatal calves after intratracheal inoculation of small volumes of fluid. Am J Vet Res 1987; 48:1589-1593. lGnabo LDB, McKellar OIL Current models in pharmacokinetics: Applications in veterinary phan'necology. Vet Res Comm 1989; 13:141-157. lGrkwood JK, Wrddowson MA Interspecies variation in the plasma halflife of oxytetracycline in relation to bodyweight. Res in Vet Sci; 48:180-183. Lam YWF, Duroux MH, Gambertoglio JG, Barrier SL, Guglielmo BJ. Effect of protein binding on serum bactericidal activities of ceftazidime and cefoperazone in healthy volunteers. Antimicrob Agents Chemother 1988; 32:298-302. Liggitt HD. Defence mechanisms in the bovine lung. Vet Clin N Am (food an prect), bovine respiratory disease edition. Philadelphia: W.B. Saunders Co. 1985; 347-366. Lillie LE. Symposium on immunization of cattle against the common diseases of the respiratory tract. The bovine respiratory disease complex. Can Vet J 1974; 15:233-242. Lillie LE. Symposium on immunization of cattle against the common diseases of the respiratory tract. The bovine respiratory disease complex. Can J Vet Res 1979; 15:233-242. Livermore DM. Permeation of B-Iactam antibiotics into Escherichia go_li, Pseudomonas aeruginosa, and other gram-negative bacteria. Rev Inf Dis 1988; 10:691-698. Loan RW. Bovine Respiratory Disease, A Symposium. Loan RW. ed. College Station, Texas: Texas A&M University Press 1984. 91 Meir TS. Value of tracheal aspirates In the diagnosis of chronic pulmonary diseases in the horse. Eq Vet J 1987; 19:463-465. Markham RJF, Wilkie BN. interaction between Pasteurella haemoMica and bovine alveolar macrophages: Cytotoxic effect on macrophages and impaired phagocytosis. Am J Vet Res 1980; 41:18-22. Mayer SE, Melmon KL, Gilmen AG. Introduction; the dynamies of drug absorption, distribution, and elimination. In: Goodman grid Giiman§ The pharmacologic_a_l basis of theremics, 6th ed. Gilmen AG, Goodman LS, Gilmen A, editors. New York: MacMillin publishing co. Inc., 1980; 1- 39. Mechor GD, Jim GK Janzen LD. Comparison of penicillin, oxytetracycline and trimethoprim- sulfedoxin in the treatment of acute undifferentiated bovine respiratory disease. Can Vet J 1988; 29:438-443. Moore RM, Walker RD, Shaw GA, Hopkins FM, Shull EP. Antileukotoxin antibody produced in the bovine lung after aerosol exposure to viable Pasteurella haemoMice. Am J Vet Res 1985; 46:1949-1952. Mosier DA, Confer AW, Panciera RJ. The evolution of vaccines for bovine pneumonic pasteurellosis. Res in Vet Sci 1989; 47:1-10. Mosier DA, Lessley BA, Confer AW, Antone SM, Gentry MJ. Chromatographic separation and characterization of Pasteurella haemoiflica cytotoxin. Am J Vet Res 1986; 47:2233-2241. Murray JF. The Normal Lung. 2nd ed. Murray JF, ed. Philadelphia: WB Saunders cc 1986: 69-82. Nakashima M, Uematsu T, Kanamaru M, Ueno K, Setoyama T, Tomono Y, Ohno T, Okano K, Morishita N. Phase 1 study of E1040, a new parenteral cephem antibiotic. J Clin Pharmacol 1989; 292144-150. Nash CH Ill, Mehta RJ, Bell C. Cephalosporin acremonium: A Beta-Iactam antibiotic-producing microbe. In: Biolggy of industrial microorganisms, Demein AL and Solomon NA, eds. Menlo Park, California: Benjamin/Cremmings publishing company, Inc., 1985; 433-448. O’Neill S, Lesperence E, Kless DJ. Rat lung lavage surfactant enhances bacterial phagocytosis and intracellular killing by alveolar macrophages. Am Rev Respir Dis 1984; 130:225-230. Panciera RJ, Corstvet RE, Confer AW, Gresham CN. Bovine pneumonic pasteurellosis: Effect of vaccination with live Pasteurella species. Am J Vet Res 1984; 45:12 2538-2542. Papich MG. Clinical pharmacology of cephalosporin antibiotics J Am Vet Med Assoc 1984; 184:344-347. Parsons RL, Beavis JP, Paddock GM, Hossack GM. Cephradine bone concentrations during total hip replacement chemotherapy In: Chemotherapy vol 1. Williams JO, Geddes AM. eds. New York: Plenum press 1976; 201-211. Peterson LR, Moody JA, Fasching CE, Gerding DA. Influence of protein binding on therapeutic efficacy of cefoperazone. Antimicrob Agents Chemother 1989; 33:566-568. 92 Philippon A, Labia R, Jacoby G. Extended-spectrum B-lactemases. Antimicrob Agents Chemother 1989; 33:1131-1136. Potgieter CND, McCracken MD, Hopkins FM, Walker RD, Guy JS. Use of fiberoptic bronchoscopy in experimental production of bovine respiratory tract disease. Am J Vet Res 1984; 45:1015-1019. Prescott JF, Baggot JD. Principles of antimicrobial drug disposition. In: Antimicrotflthergpy i_n veterinary medicine. Boston, Blackwell scientific publications 1988a; pg. 29-53. Prescott JF, Baggot JD. Principles of antimicrobial drug selection and use. In: Antimicrobial therapy in veterina_rL medicine. Boston, Blackwell scientific publications 1988b; pg. 55-69. Purdy CW, Livingston CW, Frank GH, Cummins JM, Cole NA, Loan RW. A live Pasteurella haemoMica vaccine efficacy trial. J Am Vet Med Assoc 1986; 6:589-591. Rehmtulla AJ, Thomson RG. A review of the lesions in shipping fever of cattle. Can Vet J 1981; 22:1-8. Roosendaal R, Bakker-Woudenberg iAJM, Van Der Berghe-Van Raffe M, Richel MF. Continuous versus intermittent administration of ceftazidime in experimental klebsiella pneumoniae pneumonia in normal and leukopenic rats. Antimicrob Agents Chemother 1986; 30:403-408. Rosenquist BD, Allen GK. Effect of bovine fibroblast interferon on rhinovirus infection in calves. Am J Vet Res 1190; 51:870-873. Roth JA. lmmunosuppression and immunomodulation in bovine respiratory disease. In: Bovine respiratory disease, A symposium. Loan PW, ed. College Station, Texas: Texas A&M University Press 1984:143-192. Ryan US, Grantham CJ. Metabolism of endogenous and xenobiotic substances by pulmonary vascular endothelial cells. Pharmacol Ther 1989; 42:235-0250. Sabath LD, Casey JJ, Ruch PA, Strumpf LL, Finland M. Rapid microassay for circulating nephrotoxic antibiotics. Antimicrob Agents Chemother 1969; 63-69. Sande MA, Mandell GL Chemotherapy of antimicrobial diseases In: Goodman and Gilman's the phermecolpgical basis of therapeutics. 7th ed. Gillman MA, Goodman LS, Rail 1W, Murad FD, eds. New York: Macmillan Publishing Co 1985; 1066-1094. Schentag JJ. Clinical significance of antibiotic tissue penetration. Clin Phermecokinet 1989 (Suppl. 1);16:25-31. Schuckit MA Alcohol and alcoholism. In: Harrison’s principles of Internal medicine. Eleventh ed. Braunwaid E, lsselbacher KJ, Petersdorf RG, Wilson JD, Martin JB, Fauci AS. eds. New York: McGraw-Hill Book Co. 1987; 2111. Shewen PE, Wilkie BN. Cytotoxin of Pasteurella haemoMica acting on bovine leukocytes. Inf lmm 1982; 35:91-94. Shoo MK. Experimental bovine pneumonic pasteurellosis: A review. Vet Rec 1989; 124:141- 144. 93 Short CR, Beagle RE, Arenas T, Pawlusiow J, Clarke CR. Distribution of cepharpirin into a tissue chamber implanted subcutaneously in horses. J Vet Phennacol Therapy 1987; 10:241- 247. Slocombe RF, Malark J, lngersoll R, Derksen FJ. Robinson NE. importance of neutrophils in the pathogenesis of acute pneumonic pasteurellosis in calves. Am J Vet Res 1985; 46:2253- 2258. Slocombe RF, Derksen FJ, Robinson NE. Imeractions of cold stress and Pasteurella haemoMice in the pathogenesis of pneumonic pasteurellosis in calves: method of induction and hematologic and pathologic changes. Am J Vet Res 1984; 45:1757-1763. Soback S, Ziv G. Pharmacokinetics of single doses of cefoperazone given by the intravenous and intramuscular routes to unweaned calves. Res Vet Sci 1989; 47:158-163. Soback S, Z‘w G. Pharmacokinetics and bioavailability of ceftriaxone administered intravenously and intremusculerly to calves. Am J Vet Res 1988; 49:535-538. Spratt BG, Cromie KD. Penicillin-binding proteins of Gram-negative bacteria. Rev Inf Dis 1988; 10:699-71 1. Stephens LR, Little PB, Wilkie BN, Barnum DA Infectious thromboembolic meningoencephalitis in cattle: A review. J Am Vet Med Assoc 1981; 78:378-384. Stockdale PHG, Lengford EV, Darcel CL Experimental bovine pneumonic pasteurellosis. 1. Prevention of the disease. Can J Comp Med 1979; 43:262-271. Styrt 8, Walker RD, White JC, Dehl LD, Baker JC. Granulocyte plasma membrane damage by leukotoxic supematant from Pasteurella haemoMice A1 and protection by immune serum. Can J Vet Res 1990; 54:146-150. Sykes RB. The classification and terminology of enzymes that hydrolyze B-lectam antibiotics. J inf Dis 1982; 145:762-765. Thomas LH, Gourlay RN, Wyld SG, Parsons KR, Chanter N. Evidence that blood bourne infection is involved in the pathogenesis of bovine pneumonic pasteurellosis. Vet Path 1989; 26:253-259. Thomson RG, Giika F. A brief review of pulmonary clearance of bacterial aerosols emphasizing aspects of particles relevance to veterinary medicine. Can Vet J 1974; 15:99-107. Thrupp LD. Susceptibility testing of antibiotics in liquid media. In: Antibiotics in laboratog medicine, 2nd ed. Lorian V, ed. Baltimore: Williams and Wilkins 1986; 93-150. Toothaker RD, Wright DS, Pachla LA Recent analytical methods for cephalosporins in biological fluids. Antimicrob Agents Chemother 1987; 31:1157-1163. Trigo FJ. Breeze RG, Liggitt HD, Evermann JF, Trigo E. Interaction of bovine respiratory syncytial virus and Pasteurella haemoMica in the ovine lung. Am J Vet Res 45: 1671 -1 678. Vega MV, Maheswaran SK, Lelninger JR, Ames TR. Adaptation of a colorimetric microtitration assay for quantifying Pasteurella haemoMica A1 leukotoxin and antileukotoxin. Am J Vet Res 1987; 48:1559-1564. 94 Velt HP, Farrell RL The anatomy and physiology of the bovine respiratory system relating to pulmonary disease. Comeli Vet 1978; 68:555-581. Vogeiman B, Gudmundsson S, Legett J, Turnidge J, Ebert S, Craig WA Correlation of antimicrobial pharmacokinetic parameters with therapeutic efficacy in an animal model. J Inf Des 1988; 158:831-847. Vogeiman BS, Craig WA Postantibiotic effects. J Antimicrob Chemother 1985; 15, Suppl A:37- 46. Volk WA, Benjamin DC, Kedner RJ, Parsons JT. Bacterial cell stmctures. In: Essentials of medica1microbioiogv, third edition. Philadelphia: J.B. Lippincott Company, 1986a; 244-253. Volk WA, Benjamin DC, Kedner RJ, Parsons JT. Bacterial cell structures. in: Essentials of medical microbiolpgy, third edition. Philadelphia: J.B. Lippincott Company, 1986b; 320-326. Volk WA, Benjamin DC, Kedner RJ, Parsons JT. Bacterial cell structures. In: Essentials of medical microbiology, third edition. Philadelphia: J.B. Lippincott Company, 1986c; 349-352. Vuye A, Verschraegen G, Claeys G. Plasmid-mediated B-lactamases in clinical isolates of Klebsiella mumonia and Escherichia po_Ij resistant to ceftazidime. Antimicrob Agents Chemother 1989; 33:757-761. Walker RD, Corstvet RE, Lessley BA, Panciera RJ. Study of bovine pulmonary response to fisteurella haemolvtica: Specificity of lmmunoglobulins isolated from the bovine lung. Am J Vet Res 1980; 41:1015-1023. Walker RD, Corstvet RE, Panciera RJ. Study of bovine pulmonary response to Pesteurplla haemoMieez Pulmonary macrophage response. Am J Vet Res 1980; 40:1008-1014. Walker RD, Stein GE, Hauptmen JG, MacDonald Kl-l, Budsberg SC, Rosser EJ. Serum and tissue cage fluid concentrations of ciprofioxacin after oral administration of the drug to healthy dogs. Am J Vet Res 1990; 51:896-900. Walker RD, Stein GE, Budsberg SC. Rosser EJ. MacDonald Ki-l. Serum and tissue fluid norfloxacin concentrations after oral administration of the drug to healthy dogs. Am J Vet Res 1989; 50:154-157. Weinstein L, Deikos GK, Perrin TS. Studies on the relationship of tissue fluid and blood levels of penicillin. J Lab Clin Med 1951; 38:712-71. Welling PG. Graphic methods in phannecokinetics: The basics. J Clin Phennecol 1986; 26:510-514. West JB. Pulmmrv pathophysiology, 2nd ed. Baltimore: Williams & Wilkins Co, 1982; 171- 177. Vifikse SE. Feedlot cattle pneumonias. Vet Clin N Am (food an prect), bovine respiratory disease edition. Philadelphia: W.B. Saunders Co. 1985; 289-310. Wilkie BN, Shewen PE. Bovine pneumonic pasteurellosis: a biotechnological approach to control. The Bovine Proceedings 1989; 21:52-55. 95 Wilkie lW, Fallding MH, Shewen PE, Yager JA The effect of Pasteurella haemoMica and the leukotoxin of Pasteurella haemoMica on bovine lung explants. Can J Vet Res 1990; 54:151- 156. Wrse R. Methods for evaluating the penetration of beta-Iactam antibiotics into tissues. Rev Inf Dis 1986; 8 (supp. 3):5325-5332. Wise R, Donovan IA, Ambrose NS, Allcock JE. The penetration of cefoxitin into peritoneal fluid. J Antimicrob Chemother 1981; 8:453-457. Wise R. The relevance of pharmacokinetics to In vitro models: protein binding - does it matter”? J Antimicrob Chemother 1985; 15 (supp. A):77-83. Woods GT, Mansfield ME, Webb RJ. A three-year comparison of acute respiratory disease, shrink, and weight gain in preconditioned and non-preconditioned Illinois beef calves sold at the same auction and mixed in a feedlot. Can J Comp Med 1973; 37:249. Yancey RJ, Kinney ML, Roberts BJ, Goodenough KR, Hamel JC, Ford CW. Ceftiofur sodium, a broad-spectrum cephalosporin: Evaluation in vitro and in vivo in mice. Am J Vet Res 1987; 48:1050-1053. Yates WPG. A review of Infectious Bovine Rhinotracheitis, Shipping Fever Pneumonia and viral- bacterial synergism in respiratory disease of cattle. Can J Comp Med 1982; 46:225-263. lema T, Breeze R. New technology for prevention and control of infectious bovine respiratory diseases. In: The veterinary clinics of north america, food animal practice. Bovine respiratory disease edition. Philadelphia: W.B. Saunders Co. 1985; 419-439. "11111111111111“