Tilmicosin (Micotil injection, 300 mg/ml) was supplied by Elanco Animal Health, England).

Ten calves aged one-and-a-half months, and weighing 60 to 70 kg were used in the experimental work. The calves were housed individually in adjacent boxes which had a floor area of 4 m² and a volume of 6 m³. The animals were bedded on straw. They were fed on milk replacer (free of antimicrobial substances twice daily at 09.00 and 18.00) and alfalfa with drinking water available ad libitum. The animals were shaved over jugular vein to facilitate the collection of blood samples.

The calves were allocated, randomly to two groups of 5 animals each. The first group was treated with tilmico- sin by the intravenous route and the second group was treated with tilmicosin subcutaneously. Tilmicosin solu- tion (300 mg/ml) was diluted in saline to 2.5 mg/ml and the diluted solution was administered by slow intra- venous infusion into the left jugular vein for 20 minutes at the rate of 0.5 mg/kg/min at a total dose of 10 mg/kg [12] . Each of the calves of the second group was given a single dose of 10 mg/kg of tilmicosin subcutaneously in the left dorso-lateral chest wall. All calves received tilmicosin on the same day.

Two weeks later after the intravenous and subcutaneous injection of tilmicosin, the calves were subjected to physical stress of a two-hour journey before they were inoculated with the pathogen. On day 0, approximately six hours after journey, all the calves were inoculated intratracheally with P. haemolytica type A1 (LPB 1419) [7] . A first-pass culture of P. haemolytica type A1 was inoculated onto 10 ml quantities of brain-heart-infusion broth (Oxoid), enriched with 5% fetal calf serum and incubated at 37˚C in a shaking water bath for six hours. The approximate bacterial count of each broth, after six hours of incubation, was 1 × 10⁹ cfu/ml. A polyethylene catheter (Intramedic, VEL) was inserted through the right nostril and advanced into trachea until it was 5 cm proximal to the bifurcation. The inoculum consisted of 5 ml of the six-hour culture of P. haemolytica type A1 diluted with 5 ml of sterile 0.9% sodium chloride.

The first and second groups were treated with tilmicosin (10 mg/kg body weight) by the intravenous infusion and subcutaneous injections as mentioned before. Treatment was initiated when the calves had a body tempera- ture of more than 39.5˚C and a respiration rate of more than 52/min.

The calves were examined twice a day for 10 days (5 days before and 5 days after infection) at 09.00 and 18.00. Body temperature and respiratory rate were recorded daily at 10.00.

Blood samples were obtained from the catheterized right jugular vein before and after infection. Samples were collected in heparinized tubes immediately before and at 1, 2, 4, 6, 8, 12, 24, 48, 72 and 96 hours following i.v. and s.c injections. Samples were centrifuged at 3000 rpm for 15 minutes and the obtained sera were used for the estimation of tilmicosin concentration. The serum samples were stored at −80˚C until analysis, and the assay was performed within a week of obtainment.

Bronchial secretions were collected from the calves at the same intervals of the blood samples following tilmi- cosin administration before and after infection. Bronchial secretions (0.5 ml) were collected by a tampon-device introduced into the principal bronchi through the tracheal tube [13] . The bronchial fluid was used for estimation of tilmicosin concentration.

Another blood samples were collected in vacutainer tubes (Venoject, Terumo) containing EDTA as an anti- clotting agent, once daily on the day of the inoculation (day 0) and on the following five days at 10.00 after drug administration. Total white cell counts and granulocyte/agranulocyte ratios were determined by standard me- thods.

BAL fluid samples were obtained from infected animals on day 0, 3, 4 and 5 after tilmicosin administration to infected animals. A polyethylene tube (Intramedic, VEL) with an external diameter of 4.8 mm was inserted through the right nostril and advanced into trachea and the bronchial tree until an elastic resistance felt. Fifty ml of a sterile 0.9% solution of sodium chloride at 37˚C was injected and aspirated immediately by gentle suction; approximately 75% of the infused fluid could be retrieved [7] . The neutrophil/macrophage ratio was determined in the BAL fluid. Cytological specimens were prepared by cytocentrifugation and standard WrightGiemsa stains were used.

From each BAL fluid sample, 10 ml was stored in sterile plastic tubes at 4˚C and examined for the presence of bacteria and mycoplasmas within two hours after collection. All the samples were inoculated onto three me- dia: Colombia blood agar (Oxoid), PPLO agar (Difco) enriched with 25% inactivated horse serum, 7% yeast ex- tract, 400 mg/ml ampicillin, 0.05% thallium acetate and 1% glucose and Tween 80 PPLO agar (the same en- riched and selective medium with 0.1% Tween 80) [7] . The blood plates were evaluated after 24 or 48 hours of incubation at 37˚C in a carbon dioxide-enriched atmosphere, and the mycoplasma plates were evaluated after two days, and if negative, daily until 14 days after inoculation. The bacteria were identified by the techniques described by Carter (1984) [14] .

The free tilmicosin concentrations in serum and bronchial secretions were measured by a microbiological assay technique [15] using Microccouslutea (ATCC 9431) as test organism [16] . The limit of detection of tilmicosin in serum and bronchial fluid was 4 ng/ml.

The concentrations of tilmicosin in serum and bronchial fluid were subjected to kinetic analysis and the phar- macokinetic parameters were calculated for each animal by classical methods [17] [18] . The non-compartmental pharmacokinetic parameters, volume of distribution at steady state (V_d(ss)), body clearance (Cl_(B)) and the mean residence time (MRT) were calculated according to standard methods using statistical moment theory [18] . The area under the blood concentration (C_pt) time (t) curve to infinity (AUC), and the area under the first moment curve () were calculated from the first to the last blood and bronchial sample by using the tra- pezoidal rule, and an estimate of the residual area under the curve was obtained from C_p(t*)/b, where C_p(t*) is the last measured blood and bronchial secretions concentration of tilmicosin, and b the overall elimination rate con- stant. An estimate of the residual part of the AUMC curve was obtained from C_p(t*) ´ t^*/b. The MRT, corrected for the duration of the infusion to give a corresponding value for a bolus injection, was calculated from AUC and AUMC (MRT = AUMC/AUC − T/2), where T is the infusion time. The MRT for the infusion (MRT_inf) was calculated from AUC and AUMC (MRT_inf = AUMC/AUC + T/2), where T is the infusion time (PCNONLIN, SCI, USA).

Obtained data was analyzed by analysis of variance, and the mean values were compared by Duncan’s Multiple Range test and student’s “t” test (LSD) using the SAS statistical analysis program [19] . The results are given as mean ± SD.

The slow intravenous infusion of a dilute solution of tilmicosin resulted in clinical signs suggesting acute car- diac toxicity which disappeared 25 minutes later. No side effects were observed after the subcutaneous injection.

The mean concentrations of tilmicosin in serum and bronchial secretions after tilmicosin injection in healthy and P. haemolytica-infected calves are shown in Figure 1 and Figure 2, respectively.

Values for the kinetic parameters describing the disposition of the drug are given in Table 1. Following intravenous injection of tilmicosin (10 mg/kg body weight), its concentration decreased in a biexponential man- ner that could be described in a two-compartment open model. Tilmicosin was rapidly distributed and slowly eliminated with mean half-lives of (2.70 and 2.76 hours) and (22.09 and 22.14 hours) for the distribution and elimination phases in healthy and P. haemolytica-infected calves, respectively. The volume of distribution was

^a,b,c,dMeans with the same letter in the same row are not significantly different. LSD = Least significant difference.

more than 1 L/kgindicating good penetration into tissues and bronchial secretions. The total body clearance of the drug was 0.14 and 0.12 L/kg/hin healthy and P. haemolytica-infected calves, respectively. The mean MRT corrected for a bolus injection was 2.25 and 2.20 hours; the mean MRT_inf was 25.27 and 25.46 hours and the mean V_d(ss) was 1.90 and 1.72 L/kg in healthy and P. haemolytica-infected calves, respectively.

The mean peak concentration of tilmicosin in serum (C_max, 1.33 and 1.36 mg/ml) and bronchial secretions (1.4 and 1.7 mg/ml) was reached in (7.21 and 7.15 hours) and (7.11 and 7.1 hours) after subcutaneous injection in healthy and P. haemolytica-infected calves, respectively. P. haemolytica infection significantly increased the mean peak concentration of tilmicosin in bronchial secretions as compared with pre-infection values. Table 2 and Figure 2, showed that tilmicosin was absorbed slowly from the injection site, as indicated by its long ab- sorption half-lives (t_1/2ab, 7.72 and 6.87 hours) before and after infection. The mean elimination half-life (t_1/2b) after subcutaneous injection was 24.6 and 25.85 hours indicating that the drug was eliminated slowly.

The mean body temperature of the calves is shown in Table 3. During the five days before the infection and on the day of inoculation (day 0) none of the calves had a temperature exceeding 39˚C. On the first day after in- oculation (day 1), 12 hours after the intratracheal challenge, the body temperatures of all the calves were signifi- cantly higher than before the infection and ranged between 40˚C and 40.2˚C. The body temperatures of the calves treated subcutaneously remained significantly higher than before the infection throughout the experiment. However, on day 2, 3 and 4 their body temperatures were significantly lower than on day 1.

The mean respiratory rate of the calves is shown in Table 3. During the five days preceding the infection and on the day of inoculation, the respiratory rates of the 10 calves fluctuated between 24 and 32/min. By 12 hours after the intratracheal challenge the respiratory rate of all the calves was significantly higher than those before the infection. The respiratory rates returned to the pre-inoculation values on day 4 after the infection in the group treated subcutaneously.

The means and ranges of the total WBC counts recorded in the two groups of calves on the day of the intratra- cheal challenge and on the five days after inoculation are shown in Table 4. The total WBC counts of all the calves were significantly higher on the first day after inoculation than before the intratracheal challenge. The

^a,b,cMeans with the same letter in the same row are not significantly different. LSD = Least significant difference.

^a,b,c,dMeans with the same letter in the same column measurement are not significantly different at (P > 0.0001). MSE = Minimum significant error.

counts returned to the pre-inoculation values on day 2 after the infection in the calves treated subcutaneously and on day 3 after inoculation in the slow intravenously-infused calves.

Severe granulocytosis occurred on the first day after the infection in comparison with the pre-inoculation data, but it persisted for only two days after the inoculation in intravenously treated calves.

The results of the mean Neutrophil/Macrophage (N/M) ratio in the BAL fluid are shown in Table 4. There were no significant changes in the N/M ratios in the subcutaneously treated calves.

P. haemolytica type A1 was not isolated from the BAL fluid samples before the intratracheal challenge. Fur- thermore, no other respiratory pathogens were isolated from of the calves before the infection. P. haemolytica type A1 was not isolated from BAL fluid samples of the subcutaneously-treated calves. Mycoplasma bovis was isolated from one of the five intravenously-treated calves on day 5 after inoculation. On all occasions the strains of P. haemolytica type A1 isolated were the same as that inoculated.