|Home | About | Journals | Submit | Contact Us | Français|
Pharmacokinetics of difloxacin and its distribution within the body fluids and endometrium of 6 mares were studied after intragastric (IG) administration of 5 individual doses. Difloxacin concentrations were serially measured in serum, urine, peritoneal fluid, synovial fluid, cerebrospinal fluid, and endometrium over 120 h. Bacterial susceptibility to difloxacin was determined for 174 equine pathogens over a 7-month period. Maximum serum concentration (Cmax) was 2.25 ± 0.70 μg/mL at 3.12 ± 2.63 h and Cmax after the 5th dose was 2.41 ± 0.86 μg/mL at 97.86 ± 1.45 h. The mean elimination half-life (t1/2) was 8.75 ± 2.77 h and area under the serum concentration versus time curve (AUC) was 25.13 ± 8.79 μg h/mL. Highest mean synovial fluid concentration was 1.26 ± 0.49 μg/mL at 100 h. Highest mean peritoneal fluid concentration was 1.50 ± 0.56 μg/mL at 98 h. Highest mean endometrial concentration was 0.78 ± 0.48 μg/g at 97.5 h. Mean cerebrospinal fluid concentration was 0.87 ± 0.52 μg/mL at 99 h. Highest mean urine concentration was 92.05 ± 30.35 μg/mL at 104 h. All isolates of Salmonella spp. and Pasteurella spp. were susceptible. In general, gram-negative organisms were more susceptible than gram-positives. Difloxacin appears to be safe, adequately absorbed, and well distributed to body fluids and endometrial tissues of mares and may be useful in the treatment of susceptible bacterial infections in adult horses.
La pharmacocinétique du difloxacin et sa distribution dans les fluides corporels et l’endomètre de 6 juments ont été étudiées après administration intragastrique (IG) de 5 doses individuelles. Les concentrations de difloxacin ont été mesurées à intervalles dans le sérum, l’urine, le liquide péritonéal, le liquide synovial, le liquide céphalo-rachidien, et l’endomètre durant une période de 120 h. La sensibilité au difloxacin a été déterminée pour 174 agents pathogènes équins durant une période de 7 mois. La concentration sérique maximale (Cmax) était de 2,25 ± 0,70 μg/mL à 3,12 ± 2,63 h et Cmax après la 5e dose était de 2,41 ± 0,86 μg/mL à 97,86 ± 1,45 h. La moyenne de la demi-vie d’élimination (t1/2) était de 8,75 ± 2,77 h et la surface sous la courbe de la concentration sérique en fonction du temps (AUC) était de 25,13 ± 8,79 μg h/mL. La concentration moyenne maximale atteinte dans le liquide synovial était de 1,26 ± 0,49 μg/mL à 100 h. La concentration moyenne maximale atteinte dans le liquide péritonéal était de 1,50 ± 0,56 μg/mL à 98 h. La concentration moyenne maximale atteinte dans l’endomètre était de 0,78 ± 0,48 μg/g à 97,5 h. La concentration moyenne dans le liquide céphalo-rachidien était de 0,87 ± 0,52 μg/mL à 99 h. La concentration moyenne maximale urinaire était de 92,05 ± 30,35 μg/mL à 104 h. Tous les isolats de Salmonella spp. et Pasteurella spp. étaient sensibles au difloxacin. En général, les bactéries à Gram négatif étaient plus sensibles que celles à Gram positif. Le difloxacin semble être sécuritaire, absorbé de manière adéquate, et bien distribué dans les fluides corporels et le tissu endométrial de juments et peut être utile pour le traitement d’infections causées par des bactéries sensibles chez les chevaux adultes.
(Traduit par Docteur Serge Messier)
Fluoroquinolones are a group of antimicrobials that have become widely used in veterinary medicine due to their broad-spectrum properties (1,2). These drugs exhibit bactericidal action by targeting the bacterial DNA topoisomerases II (gyrase) and IV (3,4). Principal advantages of fluoroquinolones include good oral bioavailability, bactericidal activity at low tissue concentrations, and good penetration into phagocytic cells (2,5). They have a large volume of distribution combined with low plasma protein binding, which allows them to reach tissue concentrations often higher than concurrent serum concentrations (1,2,5,6). Optimal dosing of fluoroquinolones was found to involve the administration of large dosages with long intervals between doses, as the post-antibiotic effect prevents bacterial regrowth even when blood and tissue concentrations decrease below the minimum inhibitory concentrations (MIC) (7,8).
Fluoroquinolones have been shown to be highly efficacious against many gram-negative and some gram-positive organisms (1,5), making them potentially useful in treating many serious and potentially life-threatening infections in horses. Gram-negative bacterial infections are commonly identified in clinical cases being treated in equine practice. Gram-negative bacteria (Enterobacteriaceae) have been identified as the most common bacterial group isolated from equine musculoskeletal infections (9,10). However, these same studies also found that many equine musculoskeletal infections are polymicrobial or gram-positive, with Staphylococcus sp. being the most common gram-positive organism identified. Gram-negative bacteria are also common isolates in equine pleuropneumonia (11), peritonitis (12), endometritis (13), and cholelithiasis (14).
Fluoroquinolones should be used judiciously in immature animals. High doses of difloxacin administered systemically produced cartilage lesions in immature beagles (15). In one study, articular cartilage damage was identified in foals given enrofloxacin, a fluoroquinolone antimicrobial, at doses of 10 mg/kg, PO, q24h for 8 d, beginning at 2 wk of age (16). However, enrofloxacin is widely used in mature horses and evidence of cartilage damage is rare and circumstantial (17). Adult horses given enrofloxacin (5 mg/kg body weight [BW], IV, q24h) had no signs of musculoskeletal abnormalities (18). Other adverse effects of fluroquinolones include stimulatory central nervous signs and gastrointestinal tract disturbances (19).
Difloxacin is a new fluoroquinolone antibacterial drug developed specifically for use in veterinary medicine. It has been shown to be effective in the treatment of experimentally induced pneumonic pasteurellosis in calves (20). The pharmacokinetics of difloxacin have been evaluated in dogs, and its bactericidal and inhibitory activity against small animal pathogens has been determined (21–23). Although difloxacin and other fluoroquinolones are not approved for use in horses, extra-label use of fluoroquinolones has become common in equine practices. There are currently few efficacious oral antimicrobials for use in horses, making difloxacin a potentially useful product. Pharmacokinetic and bacterial inhibitory data concerning the use of difloxacin in horses is currently lacking.
The purposes of this study were to determine the pharmacokinetic properties of difloxacin in adult mares after repeated intragastric administration, 7.5 mg/kg BW, q24h, and to measure its distribution in body fluids and endometrial tissue. Additionally, bacterial susceptibility could be determined for pathogenic bacteria isolated from equine patients at our hospital. This data could be used to develop a dosage regimen.
Five healthy Thoroughbred mares and 1 healthy Warmblood mare ranging from 7 to 19 y of age (mean, 13 y) with weights ranging from 480 to 625 kg (mean, 568 kg), were used for this experiment. They were given no other medications for a minimum of 4 wk prior to the experiment. A physical examination, complete blood count, and biochemical profile was performed for each horse prior to the experiment. During the study, basic physical parameters, such as temperature, pulse, respiration, appetite, fecal consistency, and mentation, were assessed once daily. A physical exam was also completed after the experiment. Horses were considered healthy on the basis of these findings. Bermuda grass and water were available ad libitum throughout the study. The horses were housed in stalls for the first 24 h due to frequent sampling, and then kept on pasture for the remaining portion of the study. The experimental protocol was approved by the University of Florida Animal Care and Use Committee.
Crushed difloxacin tablets (Dicural, 136 mg; Fort Dodge, Fort Dodge, Iowa, USA) suspended in 500 mL water were administered, 7.5 mg/kg BW, via nasogastric tube, without fasting, every 24 h for 5 d. The intragastric administration of difloxacin was performed after the morning feeding. Blood, synovial fluid, peritoneal fluid, cerebrospinal fluid (CSF), endometrial tissue, and urine specimens were obtained as described previously (24). Blood samples were collected from a 14-gauge intravenous jugular catheter placed in the left or right jugular vein on days 1 and 5. Each time a sample was taken, 12 mL of blood was withdrawn from the IV catheter prior to collection of the sample, and the IV catheter was flushed with heparinized saline after the sample was collected. On days 2 through 4, blood samples were obtained by jugular venipuncture just prior to and 30 min after drug administration to provide peak and trough difloxacin concentrations. All samples were taken according to a schedule (Table I). Synovial fluid samples were obtained aseptically from the antebrachiocarpal, middle carpal, and tarsocrural joints using arthrocentesis. Endometrial biopsies were obtained with an endometrial biopsy forcep (Narco Pilling, Fort Washington, Pennsylvania, USA). Cerebrospinal fluid samples were obtained at the lumbosacral space by using a 17.7 cm × 18-gauge spinal needle with stylet (Sherwood Medical, St. Louis, Missouri, USA).
Blood samples were allowed to clot and then centrifuged at 2500 × g for 15 min, and the supernatant decanted. Body fluid specimens were centrifuged at 13 000 × g for 10 min, and the supernatant decanted. All specimens were stored in polypropylene tubes at −70°C until assayed.
Concentrations of difloxacin were determined using an agar gel diffusion microbiological assay with Klebsiella pneumoniae ATCC 10031 (American Type Culture Collection, Manassas, Virginia, USA) as the assay organism (25). Known standards of difloxacin were prepared in normal equine serum on the date of the experiment and assayed simultaneously with the samples. Endometrial tissue samples were thawed, weighed, and homogenized in a tissue grinder (TenBroeck tissue grinder; Fisher Scientific, Pittsburgh, Pennsylvania, USA) with a phosphate-buffered saline solution, and centrifuged at 2000 × g for 10 min. The supernatant was assayed. The lowest limit of detection of difloxacin was 0.03 μg/mL.
Serum difloxacin concentration versus time relationship was estimated for each mare as a multiple 2 exponential plus absorption model by:
Where Cs = serum drug concentration; i = dose number (1 to 5); t = time from dose i, the initial dose; C1 and C2 = intercepts of the 2 components; e = base of Naperian logarithms; and λ1, λ2, and λ3 = slopes of the components (26). The overall elimination constant (Kel) was equated to λ2. Elimination half-life (t1/2) was calculated as the natural logarithm of 2 divided by Kel. Pharmacokinetic parameters were calculated based on non-compartmental kinetics (27). The area under the serum concentration versus time curve (AUC) was calculated from the model curve using the following equations:
The area under the moment curve (AUMC) was calculated as:
Mean residence time (MRT) was calculated from the following:
Volume of distribution based on area under the curve (Vdarea) was calculated as follows:
where F is the undetermined bioavailability.
Volume of distribution at steady state (Vdss) was determined from:
Clearance was calculated from:
Bacterial susceptibilities were determined for all equine bacterial culture specimens submitted to the microbiology laboratory of the University of Florida Veterinary Medical Teaching Hospital from November 17, 2002 to June 1, 2003. Zones of inhibition were measured from difloxacin Kirby-Bauer sensitivity disks (10 μg disks; Fort Dodge Animal Health, Fort Dodge, Iowa, USA) and susceptibility was determined using interpretive standards as described by the National Committee for Clinical Laboratory Standards (NCCLS) (28). These interpretive standards identified susceptible organisms as having zones of inhibition ≥ 21 mm, intermediate organisms with zones of 18 to 20 mm, and resistant organisms with zones ≤ 17 mm.
Minimum inhibitory concentration breakpoints for difloxacin have been determined by NCCLS (28). Quality control limits were validated once weekly with known standard organisms Enterococcus faecalis ATCC 29212, Staphylococcus aureus ATCC 29213, Escherichia coli ATCC 25922, and Pseudomonas aeruginosa ATCC 27853 (American Type Culture Collection).
No adverse effects were observed after difloxacin administration in any of the mares. After the 1st dose, maximum serum concentration (Cmax) was 2.25 ± 0.70 μg/mL at 3.12 ± 2.63 h and Cmax after the 5th dose was 2.41 ± 0.86 μg/mL at 97.86 ± 1.45 h. Mean t1/2 for difloxacin was 8.75 ± 2.77 h (Table II), with a range of 4.96 to 11.78 h.
Highest mean synovial fluid difloxacin concentration was 1.26 ± 0.49 μg/mL at 100 h (Table I). In peritoneal fluid, highest mean difloxacin concentration was 1.50 ± 0.56 μg/mL at 98 h. Highest measured mean endometrial difloxacin concentrations were 0.78 ± 0.48 μg/g at 97.5 h and 0.70 ± 0.63 μg/g at 99 h. In samples obtained from 5 of the 6 mares, mean CSF difloxacin concentration was 0.87 ± 0.52 μg/mL at 99 h. Highest mean urine difloxacin concentration was 92.05 ± 30.35 μg/mL at 104 h.
Bacterial susceptibility was determined for 174 bacterial isolates. Culture specimens were obtained from synovial fluid, peritoneal fluid, transtracheal washes, wounds, and abscesses. Difloxacin was efficacious against most of the gram-negative bacterial isolates and many gram-positive bacterial isolates. All isolates of Salmonella spp. and Pasteurella spp. were susceptible to difloxacin, with zones of inhibition ranging from 25 to 33 mm (mean, 28 mm) and 29 to 40 mm (mean, 35 mm), respectively (Table III). Thirteen of 16 E. coli isolates were susceptible, with zones of inhibition of susceptible E. coli ranging from 25 to 34 mm (mean, 29 mm). Susceptibility of gram-positive organisms to difloxacin was more variable; 9 of 10 Strep. equi subsp. equi isolates and 22 of 37 Strep. equi subsp. zooepidemicus isolates were susceptible. Zones of inhibition of susceptible Strep. equi subsp. equi isolates and Strep. equi subsp. zooepidemicus isolates ranged from 21 to 51 mm (mean, 30 mm) and 21 to 39 mm (mean, 25 mm), respectively.
A difloxacin dose of 7.5 mg/kg BW was chosen for the present study based on the oral dose recommendations of Giguere et al (2) for enrofloxacin and Haines et al (29) for orbifloxacin. An intravenous preparation was not available for use and, as a result, bioavailability could not be determined. Because of this, apparent volume of distribution was expressed as Vdss/F and clearance was expressed as clearance/F.
A microbiological assay was used in this study to determine difloxacin concentrations. Microbiological assays have also been used by other investigators to determine the response of infections to various fluoroquinolones (29–31). This bioassay does not differentiate between difloxacin and its active metabolites. It has been reported that in rabbits, difloxacin has a microbiologically active metabolite, sarafloxacin (32). This metabolite has not been reported in other species; however, it may be possible that the horse metabolizes difloxacin to sarafloxacin. When determining a dosage protocol, it is adequate to measure total antimicrobial activity without differentiation of the activity of both the parent drug and the metabolite (30).
Although the bactericidal and inhibitory activity of difloxacin has been evaluated in dogs (23,33), to our knowledge there is no published data regarding this information in horses. The MICs of difloxacin were found to be similar to those of enrofloxacin in canine bacterial isolates (34). The minimum concentration of difloxacin to inhibit 90% of E. coli canine isolates (MIC90) was ≤ 0.25 μg/mL (33). Haines et al (35) evaluated enrofloxacin using equine bacterial isolates, where the MIC90 for Enterobacteriaceae and Salmonella spp. was ≤ 0.25 μg/mL. The MIC data from these studies are similar to the trends in our Kirby-Bauer sensitivity results. The reported MIC breakpoints for difloxacin were ≤ 0.5 μg/mL for susceptible organisms, 1 to 2 μg/mL for intermediate, and ≥ 4 μg/mL for resistant organisms (28). Most gram-negative bacterial isolates were susceptible to difloxacin. In the current study, gram-positive bacterial susceptibility to difloxacin was variable. These findings correlate with previous studies (33,35).
The relationship between serum antibiotic concentrations and clinical efficacy varies between antibiotic families and is not well defined (36). Previous studies have evaluated the efficacy of fluoroquinolones against pathogenic bacteria in neutropenic rats and in ill human patients (31,37,38). In these studies, Cmax/MIC ≥ 10 and AUC/MIC ≥ 125 were suggested to be critical ratios in predicting clinical success (31,37). When these criteria are applied to pathogens with an MIC value of 0.25 μg/mL, after oral administration of difloxacin at a dose of 7.5 mg/kg BW, Cmax/MIC = 9.64 and AUC/MIC = 101. These values are near the desired ratios. Clinical doses of fluoroquinolones used in veterinary medicine often result in AUC/MIC less than the ideal values and clinical cures have been reported (39). Mean synovial fluid concentrations peaked at 1.26 μg/mL at 100 h, but remained less than serum concentrations. This concentration exceeds the MIC90 of most gram-negative organisms in previous enrofloxacin and orbifloxacin studies; however, it is lower than the MIC90 of many gram-positive organisms in these same studies (29,35). Gram-positive organisms are commonly cultured in horses with septic arthritis (10). Difloxacin may be inappropriate for the treatment of septic arthritis/tenosynovitis caused by gram-positive pathogens. Mean peritoneal fluid concentration was 1.50 μg/mL at 99 h. The majority of horses with peritonitis and positive peritoneal fluid cultures have polymicrobial infections, with E. coli being the most common isolate from such samples (12). Difloxacin may be efficacious in the treatment of peritonitis originating from sensitive gram-negative organisms.
Mean urine difloxacin concentrations were 20 to 61 times higher than concurrent serum concentrations, making difloxacin potentially useful in the treatment of urinary tract infections. This is consistent with the renal excretion of difloxacin and its active metabolites. These concentrations are lower than those reported in studies evaluating other fluoroquinolones in horses (2,29,35). Mean CSF concentration at 99 h was approximately 43% of the concurrent mean serum concentration. These concentrations are representative of horses with apparently normal blood-brain barriers. Horses with inflammation of the central nervous system may have altered penetration of difloxacin into the CSF. The highest measured mean concentration of difloxacin attained in endometrial tissue was 0.78 μg/g. When using bacterial MICs from the enrofloxacin study by Haines et al (35), the MIC90 for both E. coli and Klebsiella spp. exceeds, but does not meet the MIC90 for Streptococcus zooepidemicus. The clinical response of endometritis to systemic antibiotics has not been critically evaluated (13).
Difloxacin was well distributed throughout the body fluids and endometrial tissue. Difloxacin displayed a long t1/2 (8.75 h), similar to that in dogs (21,22). This t1/2 is also similar to that of other fluoroquinolones in the horse (2,29,35). During the first 24 h, the mean peak serum concentration of difloxacin was 1.96 μg/mL, occurring 140 min after intragastric administration. While this concentration is similar to those seen in other fluoroquinolones, the time of peak serum concentration appears to be later. This compares to values reported for orbifloxacin of 2.41 μg/mL at 90 min (29), and for enrofloxacin of 1.85 μg/mL at 45 to 60 min after dosing (35,40). Absorption of difloxacin appeared to be variable between horses, with a coefficient of variation for AUC of 35%. This is similar to the absorptive variability of enrofloxacin with a coefficient of variation of 76% in one study (40) and 29% in another (35).
Based on the results of this study, difloxacin given orally, 7.5 mg/kg BW, q24h, to non-fasted adult horses appears to be safe, adequately absorbed, and well distributed throughout body fluids and endometrial tissues. Using Cmax/MIC ≥ 10 and AUC/MIC ≥ 125 as criteria for evaluating efficacy of fluoroquinolones (31,37), this dose may be appropriate for treating pathogens with an MIC value ≤ 0.25 μg/mL. To the authors’ knowledge, these criteria have not been substantiated in the horse and warrant further investigation. The determination of specific MICs of difloxacin to various equine pathogens may be useful in further evaluating its efficacy. It is unknown whether administration of difloxacin to young growing horses should be avoided until more studies have evaluated potential side effects on articular cartilage.
The authors thank Dr. Sheila Morris for technical assistance, An Nguyen and Jeffery O’Kelley for assistance with microbiological data, and Fort Dodge Animal Health for supplying Dicural.
This study was supported by UFCVM Consolidated Research Development Award Competition grant number 7254044-12.