Pharmacokinetics, urinary excretion and milk penetration of levofloxacin in lactating goats

Levofloxacin, a recently introduced third-generation fluoroquinolone, is the l-isomer ofloxacin and possesses excellent activity against Gram-positive, Gram-negative and anaerobic bacteria (North et al., 1998). Compared with other fluoroquinolones (FQs), it also has more pronounced bactericidal activity against organisms such as Pseudomonas, Enterobacteriaceae and Klebsiella spp. (Klesel et al., 1995). Several species of staphylococci, streptococci including Streptococcus pneumoniae, bacteroides, clostridium, haemophilus, moraxella, legionella, mycoplasma and chlamydia are susceptible to levofloxacin (Langtry & Lamb, 1998). The bactericidal effect of levofloxacin is achieved through reversible binding to DNA gyrase and subsequent inhibition of bacterial DNA replication and transcription (Fu et al., 1992). Levofloxacin distributes well to target body tissues and fluids in the respiratory tract, skin, urine and prostrate, and its uptake by cells makes it suitable for use against intracellular pathogens. However, it penetrates poorly into the central nervous system (Langtry & Lamb, 1998). FQs act by a concentration-dependent killing mechanism, whereby the optimal effect is attained by the administration of high doses over a short period of time (Drusano et al., 1993). This concentration-dependent killing profile is associated with a relatively prolonged postantibiotic effect (Aliabadi & Lees, 2001). The drug undergoes a limited metabolism in rats and human (Langtry & Lamb, 1998) and is primarily excreted by kidney mainly as active drug. Inactive metabolites (N-oxide and demethyl metabolites) represent <5% of the total dose (Hurst et al., 2002). The pharmacokinetics of levofloxacin has been fully investigated in humans (Chulavatnatol et al., 1999), rabbits (Destache et al., 2001), cats (Albarellos et al., 2005) and calves (Dumka & Srivastava, 2006, 2007). However, there is no information available on the pharmacokinetics of levofloxacin in goats. In view of the marked species variation in the kinetic data of antimicrobial drugs, the present study was undertaken to determine the pharmacokinetics, urinary excretion and milk penetration of levofloxacin following single intravenous (i.v.) and intramuscular (i.m.) administration in lactating goats. Tavanic® [100 mL vial of solution of levofloxacin hemihydrate equivalent to 500 mg (5 mg/mL) levofloxacin] was purchased from Aventis, Frankfurt, Germany and Mueller–Hinton agar from Mast Group Ltd., Merseyside, UK. Where Cp is the concentration of drug in the plasma at time t; A is the intercept of the distribution phase with the concentration axis expressed as μg/mL; B is the intercept of the elimination phase with the concentration axis expressed as μg/mL; α is the distribution rate constant expressed in units of reciprocal time (/h); β is the elimination rate constant expressed in units of reciprocal time (/h) and e is the natural logarithm base. The i.m. data were analysed by adopting a one-compartment open model. This program also calculated noncompartmental parameters using the statistical moment theory (Gibaldi & Perrier, 1982). The maximum plasma concentration (Cmax) and time of maximum plasma concentration (Tmax) were taken directly from the curve. The area under plasma concentration–time curve (AUC) and the area under the first moment curve (AUMC) were calculated by using the method of trapezoids, and extrapolation to infinity was performed. The milk time–concentration data were analysed with R Strip programme using the drug concentration at each sampling time interval. The extent of drug penetration from the blood into the milk was expressed as the ratios of AUCmilk/AUCplasma and Cmax-milk/Cmax-plasma (Ziv et al., 1995). Pharmacodynamic efficacy of levofloxacin was determined by calculating the Cmax/MIC90 and AUC24/MIC90 ratios following i.m. administration using the respective mean MIC value for susceptible organisms. There are no studies reporting MIC90 values for levofloxacin for bacteria isolated from goats. So, to calculate the PK/PD efficacy predictors, the MIC90 value (MIC ≤ 0.17 μg/mL) was used based on other veterinary FQs against sensitive strains isolated from field of veterinary importance (Watts et al., 1997). The mean plasma pharmacokinetic variables for levofloxacin were statistically compared by nonparametric analysis, using the Mann–Whitney test and instant version 3.00 (GraphPad Software). The mean values were considered significantly different at P < 0.05, 0.01 and 0.001. Clinical examination of all animals before and after each trial did not reveal any abnormalities. No local or systemic adverse reaction to levofloxacin occurred after i.v. or i.m. administration. The mean plasma, milk and urine concentration–time profiles of levofloxacin following single i.v. and i.m. administrations of 4 mg/kg b.w. are presented graphically in Fig. 1. The mean ± SD values of plasma and milk pharmacokinetic parameters estimated from the curve fitting after i.v. or i.m. administration are shown in Table 1. Mean ± SD of plasma (^i.v.,•i.m.), urine (Δ i.v.,i.m.) and milk (oi.v.,•i.m.) concentrations of levofloxacin in lactating goats after i.v. and i.m. administration of 4 mg/kg body weight (n = 6). As there is no report of significant active metabolites in rats or human beings, the application of the microbiological assay for measuring levofloxacin concentration is suitable. Plasma levofloxacin disposition curves after i.v. injection were best fit to an open bicompartmental model in all the animals, which is in accordance with the results reported for calves (Dumka & Srivastava, 2007). The K12/K21 ratio was 1.13 indicating a faster drug transportation rate from the central to the peripheral compartment than redistribution from the peripheral to the central compartment. The Vdss for levofloxacin was 0.73 L/kg in lactating goats indicating a relatively wide distribution after i.v. administration and it was consistent with those that reported for moxifloxacin by Fernandez-Varon et al. (2006) in lactating goats (0.79 L/kg). Levofloxacin’s clearance in lactating goats was (0.18 L/h/kg) similar to the value reported in calves (0.19 L/h/kg, Dumka & Srivastava, 2007). The extent of renal elimination varies across the FQs. Levofloxacin is eliminated primarily by the kidney, with the renal clearance exceeding creatinine clearance by approximately 60% (Martinez et al., 2006) suggesting the involvement of both glomerular filtration and tubular section (Okazaki et al., 1991). The finding was confirmed by a 24–35% decrease in renal clearance of levofloxacin following doses of probenecid (Aminimanizani et al., 2001). Elimination half-lives were (2.95 and 3.64 h) close to those reported for levofloxacin in calves after i.v. and i.m. dosing respectively (Dumka & Srivastava, 2006, 2007). The drug was eliminated from plasma after i.m. treatment at a significantly slower rate than after i.v. treatment, suggesting the presence of a ‘flip–flop’ effect at least after i.m. administration (Toutain & Bousquet-Melou, 2004). In that model, the last phase of the curve is determined by the absorption rate constant and not by the apparent elimination constant, because the rate of absorption is a limiting factor for the elimination process. The peak plasma level of levofloxacin attained in the present study was approximately 19-fold higher than the MIC of levofloxacin and the drug was detected above the minimum therapeutic plasma level up to 24 h of administration. Similar to our findings, a peak plasma concentration of 3.07 μg/mL was attained after single i.m. injection of levofloxacin in calves (Dumka & Srivastava, 2006). The systemic bioavailability of levofloxacin in lactating goats after i.m. administration was 85% indicating good absorption of the drug from that injection site. This value was lower than that reported for other FQs in nonlactating goats such as moxifloxacin (97%; Fernandez-Varon et al., 2006), difloxacin (90%; Marin et al., 2007) and danofloxacin (110%; Aliabadi & Lees, 2001). This variation may be attributed to the more rapid elimination of fluoroquinolone in lactating animals (Petracca et al., 1993). The mean binding of levofloxacin to the plasma proteins of goats (22%) was in accordance with the corresponding values of 24% in human (Langtry & Lamb, 1998). Nevertheless, it was relatively lower to that reported (17%) in calves (Dumka & Srivastava, 2006). The extent of protein binding in milk was higher (37%) than in plasma. Similar finding has been reported for norfloxacin in lactating cows (Gips & Soback, 1999). Levofloxacin, as with several other FQs, is amphoteric because of the presence of a carboxylic acid and one or more basic amine functional groups. Passive diffusion across biological membranes is a function of fluoroquinolone lipophilicity relative to the pKa values of the two ionizable moieties. The good penetration of levofloxacin from the blood into the goat’s milk at pH 6.5–6.7 was predicted. From this data, levofloxacin could have been successful against susceptible mastitic pathogens in goats after parenteral administration. Levofloxacin urine concentrations were (10–18 times) much higher than those of plasma and milk and they could be detected in urine till 36 h postinjections by both the routes. The concentration of levofloxacin-equivalent inhibitory units in the urine was very high, by approximately 30 times the MIC90 even 24 h after administration. High urinary concentration of danofloxacin (58.58 μg/mL) has also been reported after i.v. doses of 1.25 mg/kg in goats (Atef et al., 2001). Approximately 55% of the microbiological activity of the administered drug was recovered in the urine of goats within 24 h. These findings suggest that levofloxacin may be an appropriate drug for treating urinary tract infections in goats. It has been established that for concentration-dependant FQs, the AUC0–24/MIC90 is the most important efficacy predictor with the rate of clinical cure being >80%, when this ratio is higher than 100–125 (Lode et al., 1998). A second predictor of efficacy for concentration dependent antibiotic is the ratio Cmax/MIC90, considering that values above 8–10 would lead to better clinical results (Dudley, 1991). It is now accepted that high Cmax/MIC90 values are necessary to avoid bacterial resistance emergence (Walker, 2000). MIC90 data of levofloxacin against caprine bacterial isolates have not been reported up to now. So if we take into account, MICs of other veterinary FQs against sensitive strains of different micro-organisms isolated from field of veterinary importance (Watts et al., 1997) and using the surrogate ratios ratios AUC0–24/MIC90 (124.5) and Cmax/MIC90 (18.6), levofloxacin could be effective by the i.m. route at 4 mg/kg against bacterial isolates with MIC ≤ 0.17 μg/mL.