A Comprehensive Model for Enrofloxacin to Ciprofloxacin Transformation and Disposition in Dog

A Comprehensive Model for Enrofloxacin to Ciprofloxacin Transformation and Disposition in Dog C. C. CESTER

AND

P. L. TOUTAINX

Received August 20, 1996, from the Unite´ Associe´e INRA de Physiopathologie & Toxicologie Expe´rimentales, Ecole Nationale Ve´te´rinaire, 23 chemin des Capelles, F-31076 Toulouse Cedex, France. Final revised manuscript received June 16, 1997. Accepted for publication June 18, 1997X. Abstract 0 The pharmacokinetics of enrofloxacin and ciprofloxacin, its major active metabolite, were determined in dog after oral and intravenous administrations of enrofloxacin and intravenous infusion of ciprofloxacin. A comprehensive model of enrofloxacin and ciprofloxacin disposition was constructed to investigate the extent of enrofloxacin to ciprofloxacin transformation and the influence of the hepatic first-pass effect on the parent compound oral bioavailability. Plasma levels were measured using a validated HPLC method. Enrofloxacin and ciprofloxacin plasma concentration data were fitted simultaneously using a set of differential equations describing a six-compartment model (two compartments for each analyte, one for the liver, and one for the intestinal tract); it was assumed that only a fraction of enrofloxacin was metabolized to ciprofloxacin and that this conversion only occurred in the liver. The fitted parameters obtained from the model were used to calculate plasma clearances (0.729 ± 0.212 L/h/kg for enrofloxacin, 0.468 ± 0.094 L/h/kg for ciprofloxacin), distribution volumes (2.45 ± 0.49 L/kg for enrofloxacin, 1.92 ± 0.33 L/kg for ciprofloxacin), mean residence times (3.47 ± 0.78 h for enrofloxacin, 4.20 ± 0.82 h for ciprofloxacin), and the fractions of enrofloxacin metabolized to ciprofloxacin after intravenous and oral administrations of enrofloxacin. It was shown that enrofloxacin was largely metabolized to ciprofloxacin and that the fractions of metabolized enrofloxacin were similar after intravenous and oral administrations of enrofloxacin (40.44 ± 10.08 and 40.17 ± 8.33%, respectively), the hepatic first-pass effect being low (7.15 ± 1.99%).

Introduction Enrofloxacin is a fluoroquinolone with an extended spectrum of anti-microbiological activity which is specifically marketed for veterinary medicine. In several species including horse,1 cattle,2 and dog,3 it has been shown that enrofloxacin is de-ethylated into ciprofloxacin, a more active antimicrobial agent against certain bacteria than enrofloxacin4,5 (Figure 1). This has led some authors to suggest that enrofloxacin is mainly a prodrug of the more potent ciprofloxacin, a considerable part of the antimicrobial activity of enrofloxacin being due to its main metabolite.3 In fact the extent of the actual metabolism of enrofloxacin to ciprofloxacin has never been measured and the aforementioned conclusions are only based on the indirect evidence that plasma ciprofloxacin concentration, after enrofloxacin administration, accounted for a relevant percentage of the total plasma drug concentration. It must be realized that plasma ciprofloxacin concentration after enrofloxacin administration relies not only on the extent of its formation from enrofloxacin but also on its own volume of distribution and clearance. Thus, the drug to metabolite ratio is not a univocal and accurate endpoint for discussion of the extent of a metabolic pathway. X

Abstract published in Advance ACS Abstracts, August 15, 1997.

1148 / Journal of Pharmaceutical Sciences Vol. 86, No. 10, October 1997

Figure 1sStructures of enrofloxacin and ciprofloxacin.

The purpose of the present experiment was to develop a comprehensive model of enrofloxacin/ciprofloxacin disposition in dog, in order to answer certain basic questions such as the actual extent of enrofloxacin to ciprofloxacin transformation and the influence of the hepatic first-pass effect on enrofloxacin systemic oral bioavailability.

Materials and Methods AnimalssExperiments were carried out in eight beagle dogs (four males and four females), 2-5 years old and weighing 12-20 kg at the beginning of the trial. They were fed with a commercial dog food and were offered water ad libitum. Experimental Design and Drug AdministrationsA parallel design was performed for the present investigation because it was part of a larger study and a crossover design was not the better design for this purpose. First, enrofloxacin (Baytril 15 mg and Baytril 50 mg Bayer AG, Leverkusen, Germany) at a dose of 5 mg/kg was administered orally to all dogs. The dose actually administered by oral route differed slightly from the nominal dose depending on the weight of the dog. The dose was adjusted to the body weight of the dog using the two dosage strengths of the enrofloxacin tablets; the actual average oral dose was 4.91 ( 0.36 mg/kg. Three weeks after the oral administration, the dogs then received an intravenous administration of 5 mg/kg enrofloxacin (Baytril 2.5% Bayer AG, Leverkusen, Germany) administered by means of a catheter inserted into one of the cephalic veins. An intravenous infusion of ciprofloxacin (Ciflox, Bayer Pharma, Puteaux, France) at a dose of 5 mg/kg was given after a further washout period of 3 weeks. For all administrations, the dogs received a meal at 3:00-4:00 p.m. on the day before the trial and then remained unfed until the antibiotics were administered.

S0022-3549(96)00346-2 CCC: $14.00

© 1997, American Chemical Society and American Pharmaceutical Association

Figure 2sSix-compartment model (two compartments for each analyte, one for the liver, and one for the intestinal tract) used to describe the enrofloxacin disposition. Kij, first rate constant of transfer of the drug from compartment i to compartment j; Ki0, rate constant of elimination from compartment i. Sample Collectionsblood samples (6 mL) were collected by direct venipuncture from the jugular veins, and the plasma was separated by centrifugation (1 400g, 10 min at 6-9°C), within 30 min of sample collection. Aliquot fractions were stored at -20 °C until assayed. The sampling times for oral administration were before, 5, 10, 15, 30, and 45 min, and 1, 2, 3, 4, 6, 8, 10, 12, 24, 32, 48, 54, and 72 h after the antibiotic administration. For the intravenous route, blood samples (6 mL) were collected before, 2, 4, 8, 15, 30, and 45 min, and 1, 2, 3, 4, 6, 8, 10, 12, 24, 32, 48, 54, and 72 h after the enrofloxacin administration. The sampling times after the intravenous infusion of ciprofloxacin were as follows: before, 3, 5, and 8 min after the beginning of the ciprofloxacin infusion, and 1, 2, 4, 8, 15 30, 45, 60, and 90 min and 2, 3, 4, 6, 8, 10, 24, 32, 48, 54, and 72 h after the end of the infusion. The dogs were fed after the 12, 32, 54, and 72 h blood samplings (iv and po enrofloxacin administrations) and after the 10, 32, 54, and 72 h blood sampling (iv ciprofloxacin infusion). Analytical Assaysthe enrofloxacin and ciprofloxacin plasma concentrations were determined using a validated HPLC highperformance liquid-chromatographic method. The criteria used for the validation were the current standards.6 Analytical assays were performed within the 4 months after the end of animal phase. Plasma enrofloxacin and/or ciprofloxacin concentrations were determined by HPLC method using ofloxacin as internal standard. The tested fluoroquinolones and the internal standard were extracted with the same liquid-liquid extraction method using chloroform as previously described.7 The HPLC apparatus consisted of a pump system equipped with an automatic injector, variable wavelength detector, and a recording software (Millipore Corp. Waters Chromatography, Milford, MA). Separation was accomplished for enrofloxacin and ciprofloxacin by a reverse-phase column (Novapak C18, 4 µm, 250 × 4.6 mm, Millipore Corp. Waters Chromatography, Milford, MA). The mobile phase consisted of a mixture of methanol, acetonitrile, pH 2.7 buffer, acetic acid, and triethylamine (10/2.5/86.5/1/20, v/v/v/v/mM). The pH 2.7 buffer was a 0.4% aqueous solution of diammonium hydrogen phosphate and tetrabutylammonium hydrogen sulfate. A flow rate of 1 mL/min was used for all the chromatographic procedures. Detection was by UV absorption at 279 nm for enrofloxacin and ciprofloxacin. The retention times of ofloxacin, ciprofloxacin, and enrofloxacin were 11.6, 21.6, and 25.8 min, respectively. When ciprofloxacin was assayed alone (after the iv infusion), the retention times were decreased to 10.2 and 18.2 min for ofloxacin and ciprofloxacin, respectively. Eight standard concentration levels ranging from 0.02 to 5.0 µg/mL were used for the calibration curves for enrofloxacin and ciprofloxacin. Quality control specimens were

coanalyzed: three concentration levels were assayed in duplicate at each analytical run. The validated limits of quantification of the methods were 0.02 µg/mL for enrofloxacin and ciprofloxacin. The extraction recoveries were greater than 80% and repeatability and reproducibility as measured by their coefficients of variation were lower than 10%. Data Analysissusing the Akaike information criterion (AIC),8 the compartmental model depicted in Figure 2 was selected among other different models to simultaneously fit plasma enrofloxacin and ciprofloxacin molar concentrations, obtained after iv and po enrofloxacin administration, and plasma ciprofloxacin molar concentrations obtained after iv ciprofloxacin administration. The model assumed that only a fraction of enrofloxacin (compartment 1) was metabolized to ciprofloxacin (compartment 3) and that transformation only occurred in the liver (compartment 2). A liver compartment was materialized so that enrofloxacin absorption (Fabs), the fraction of enrofloxacin that survived an hepatic first-pass effect (Fh), and the overall enrofloxacin bioavailability (F) could be evaluated separately. The model was assumed linear and the Kij were the first order rate constants of absorption, transfer or conversion from compartment i to compartment j. F was obtained from eq 1:

F ) FabsFh

(1)

Fabs was calculated as

Fabs )

K62 K62 + K60

(2)

where K62 is the rate constant of enrofloxacin absorption from the digestive tract and K60, the rate constant corresponding to the unabsorbed enrofloxacin from the digestive tract. Fh was calculated as

Fh )

K21 K21 + K23

(3)

where K21, is the rate constant of enrofloxacin transfer from liver to its central compartment and K23, the rate constant of enrofloxacin to ciprofloxacin conversion. The liver extraction ratio (Eh) was calculated as

Eh ) 1 - Fh

(4)

Journal of Pharmaceutical Sciences / 1149 Vol. 86, No. 10, October 1997

Enrofloxacin and ciprofloxacin had peripheral compartments (compartments 4 and 5, respectively). The model was constructed from a set of differential equations. A fifth-order Runge-Kutta method with variable step size was used to solve the model numerically. Initial estimates were obtained by separately analyzing enrofloxacin and ciprofloxacin (iv, po) using classical equations.9 The parameters were refined by the REVOL minimizing algorithm.10 The goodness of fit of the described model was estimated using least squares criteria weighted with 1/yˆ 2, yˆ being the fitted value. Plasma enrofloxacin and ciprofloxacin clearances were obtained from the fitted parameters. Enrofloxacin clearance was calculated according to the general treatment to be used for linear mammillary models which includes a first-order elimination from compartments other than the central compartment.11 Enrofloxacin clearance was

(

ENRO CLENRO,iv ) V1 K10 +

K23K12 K21 + K23

)

(5)

lated by integrating the model with the fitted parameters and by applying the trapezoidal rule to the fitted ciprofloxacin concentrations. The mean formation times (MFT) of ciprofloxacin from iv and po enrofloxacin administrations were obtained by eqs 13 and 14, respectively: CIPRO CIPRO CIPRO ) MRTENRO,iv - MRTCIPRO,iv MFTENRO,iv

(13)

CIPRO CIPRO CIPRO MFTENRO,po ) MRTENRO,po - MRTCIPRO,iv

(14)

The fraction (fm1) of the enrofloxacin dose metabolized to ciprofloxacin after an iv enrofloxacin administration was obtained from the total enrofloxacin clearance (eq 5) and the fraction of the enrofloxacin clearance not transformed to ciprofloxacin, i.e. K10V1, thus

fm1 )

K23K12 K10(K21 + K23) + K23K12

(15)

where V1 is the volume of the central enrofloxacin compartment, K10 is the rate constant of irreversible elimination of enrofloxacin from the enrofloxacin central compartment, K12 is the exchange rate constant between the central enrofloxacin compartment and the enrofloxacin liver compartment, and K21 and K23 are defined above. For ciprofloxacin, clearance was

The fraction (fm2) of the enrofloxacin dose metabolized to ciprofloxacin after a po enrofloxacin administration was the sum of the enrofloxacin fraction which was absorbed and directly transformed to ciprofloxacin by the first-pass effect and of the fraction (fm1) of the bioavailable enrofloxacin which was transformed to ciprofloxacin (eq 16)

CIPRO CLCIPRO,iv ) V3K30

fm2 ) fm1F + (1 - Fh)Fabs

(6)

where V3 is the volume of ciprofloxacin central compartment and K30, the rate constant of irreversible ciprofloxacin elimination. The mean residence time (MRT) of enrofloxacin after iv enrofloxacin administration was

(

ENRO MRTENRO,iv ) 1+

)(

K12 K14 K23K12 + / K10 + K23 + K21 K41 K23 + K21

)

(7)

where K14 and K41 are the rate constants of transfer between the central and peripheral compartments of enrofloxacin and K12, K23, K21, and K10 are defined above. The mean residence time of ciprofloxacin after iv ciprofloxacin administration was CIPRO MRTCIPRO,iv

K35 + K53 ) K53K30

(8)

where K35 and K53 are the rate constants of exchange between the central and peripheral compartments of ciprofloxacin; K30 is defined above. The steady state volume of distribution (Vss) for enrofloxacin and ciprofloxacin was obtained by multiplying the corresponding clearance and mean residence time, i.e. ENRO ENRO ENRO VssENRO,iv ) CLENRO,iv MRTENRO,iv

(9)

and CIPRO CIPRO CIPRO VssCIPRO,iv ) CLCIPRO,iv MRTCIPRO,iv

(10)

ENRO CIPRO ENRO CIPRO with CLENRO,iv , CLCIPRO,iv , MRTENRO,iv , and MRTCIPRO,iv as defined above (eqs 5, 6, 7, and 8, respectively). The mean absorption time (MAT) of enrofloxacin after po enrofloxacin administration was

ENRO MATENRO,po ) 1/(K60 + K62)

(11)

where K60 and K62 are defined above (see eq 2). The mean residence time of enrofloxacin after po enrofloxacin administration was equal to ENRO ENRO ENRO MRTENRO,po ) MRTENRO,iv + MATENRO,po

(12)

ENRO ENRO with MRTENRO,iv , given by eq 7 and MATENRO,po given by eq 11. The mean residence time of ciprofloxacin after iv and po enrofCIPRO CIPRO loxacin administration (i.e. MRTENRO,iv and MRTENRO,po ) was calcu-

1150 / Journal of Pharmaceutical Sciences Vol. 86, No. 10, October 1997

(16)

with F, Fabs and Fh given in equations 1, 2, and 3 respectively. As the actual oral dose differed slightly from the nominal dose, depending on the dog’s weight, the relevant pharmacokinetic parameters were adjusted for actual dose. Statistical AnalysissThe results are reported as mean ( standard deviation (SD). For the comparison of enrofloxacin and ciprofloxacin HPLC parameters, a statistical analysis was carried out using a three factor analysis of variance (ANOVA). The three factors were: drug effect (fixed, enrofloxacin versus ciprofloxacin), sex effect (fixed) and dogs effect nested in sequence effect (random, 4 dogs in each sex). The level of significance was 0.05.

Results The mean plasma concentrations versus time profile curves for enrofloxacin and ciprofloxacin after a single iv or po nominal dose of 5 mg/kg of enrofloxacin are shown in parts A and B of Figure 3, respectively. The mean plasma concentration versus time profile for ciprofloxacin after a short iv infusion of ciprofloxacin and after iv and po bolus enrofloxacin administrations are shown in Figure 3C. All the data sets were successfully fitted using the model in Figure 2. Parts A and B of Figure 4 show the observed data points and fitted curve for both enrofloxacin and ciprofloxacin after iv or po enrofloxacin administrations for a representative dog, respectively. Figure 4C shows the observed data points and fitted curve for ciprofloxacin after an iv ciprofloxacin infusion, for the same representative dog. The estimated parameters of the model are given in Table 1. The volumes of the central compartment for enrofloxacin (V1 ) 0.97 ( 0.11 L/kg) and ciprofloxacin (V3 ) 0.13 ( 0.03 L/kg) were significantly different (ANOVA, p < 0.001). Other parameters having a primary physiological meaning were calculated from the estimated rate constants and volumes of distribution of the model (Table 2). The plasma clearance was significantly higher (ANOVA, p < 0.05) for enrofloxacin (0.73 ( 0.21 L/h/kg) than for ciprofloxacin (0.47 ( 0.09 L/h/kg). The steady state volume of distribution was significantly but marginally higher for enrofloxacin (2.45 ( 0.49 L/kg) than for ciprofloxacin (1.92 ( 0.33 L/kg) (ANOVA, p < 0.01). No sex effect was evidenced for the tested parameters except for the steady state volume of distribution of enrofloxacin and ciprofloxacin (ANOVA, p < 0.01). Finally, the mean residence time of iv administered ciprofloxacin was slightly but not signifi-

Figure 3sSemilogarithmic plot of the mean plasma concentrations versus time of enrofloxacin (9) and ciprofloxacin (0), after iv (A) and po (B) administrations of enrofloxacin (5 mg/kg) in eight dogs. (C) Semilogarithmic plot of the mean plasma concentrations versus time of ciprofloxacin, after iv (0) and po (O) administrations of enrofloxcacin and iv infusion (9) of ciprofloxacin (5 mg/kg) in eight dogs.

cantly higher than for iv administered enrofloxacin (4.20 ( 0.82 h versus 3.47 ( 0.78 h, ANOVA, p > 0.05). The mean residence time of ciprofloxacin after an enrofloxacin iv administration was 7.73 ( 1.20 h, and the mean formation time

of ciprofloxacin from an enrofloxacin iv administration was 3.52 ( 0.78 h (Table 3). After oral enrofloxacin administration, the mean absorption time of enrofloxacin was 1.42 ( 0.85 h. The MRT was 4.89 (

Journal of Pharmaceutical Sciences / 1151 Vol. 86, No. 10, October 1997

Figure 4sSemilogarithmic plot of the fitted and observed plasma concentrations versus time of enrofloxacin (observed, 9) and ciprofloxacin (observed, 0), after iv (A) and po (B) administrations of enrofloxacin (5 mg/kg) in a representiative dog. (C) Semilogarithmic plot of the fitted and observed (0) plasma concentrations versus time of ciprofloxacin after iv infusion of ciprofloxacin (5 mg/kg) in a representative dog.

1.25 h for enrofloxacin and 8.66 ( 1.76 h for ciprofloxacin, after a po enrofloxacin administration. The mean formation time of ciprofloxacin after an oral enrofloxacin administration 1152 / Journal of Pharmaceutical Sciences Vol. 86, No. 10, October 1997

was 4.45 ( 1.26 h. The absorbed enrofloxacin fraction (Fabs) was high (89.9 ( 8.0%) and the absorbed enrofloxacin fraction which escaped a single passage though the liver (Fh) was large

Table 1sEstimated First-Order Rate Constants (Kij) and Volumes of Distribution for Enrofloxacin and Ciprofloxacin Obtained by the Simultaneous Fitting of Plasma Concentrations of Enrofloxacin and Ciprofloxacin after Iv Enrofloxacin Administration, Po Enrofloxacin Administration, and Iv Ciprofloxacin Administration Dogsb Parametersa

1 (M)

2 (F)

3 (M)

4 (F)

5 (M)

6 (F)

7 (M)

8 (F)

Mean ± SD

K10 K12 (h-1) K21 (h-1) K23 (h-1) K62 (h-1) K60 (h-1) K30 (h-1) K14 (h-1) K41 (h-1) K35 (h-1) K53 (h-1) V1 (L/kg) V3 (L/kg)

0.319 2.878 7.943 0.548 0.543 0.025 3.721 6.021 6.110 20.315 1.183 1.094 0.126

0.550 4.492 13.908 0.984 1.391 0.086 3.433 4.539 3.565 13.953 1.645 0.894 0.166

0.340 3.992 9.986 0.658 0.426 0.030 2.737 4.184 3.104 20.733 1.896 1.013 0.161

0.330 4.820 9.733 0.573 1.348 0.034 3.009 4.775 6.352 16.927 1.589 1.063 0.171

0.434 2.857 7.929 0.867 0.953 0.290 4.800 4.193 4.010 24.885 1.617 0.937 0.127

0.441 4.401 11.423 0.647 0.259 0.069 3.341 2.796 5.653 16.652 1.184 1.012 0.100

0.478 12.068 16.404 1.142 0.719 0.104 4.155 4.678 2.500 34.058 1.608 0.973 0.107

0.682 2.541 10.244 1.224 0.998 0.049 3.442 6.803 4.490 16.746 1.163 0.746 0.106

0.447 ± 0.125 4.756 ± 3.077 10.946 ± 2.923 0.830 ± 0.264 0.830 ± 0.416 0.086 ± 0.087 3.580 ± 0.652 4.748 ± 1.215 4.473 ± 1.435 20.534 ± 6.411 1.486 ± 0.274 0.966 ± 0.110 0.133 ± 0.029

(h-1)

a K , estimated first-order rate constants of transfer from compartment i to compartment j (see Figure 2 for the different compartments); V and V , volumes of the ij 1 3 central compartment of enrofloxacin and ciprofloxacin, respectively. b (M) male beagle dog and (F) female beagle dog.

Table 2sDerived Pharmacokinetic Parameters for Enrofloxacin and Ciprofloxacin As Obtained after an Iv Administration of Enrofloxacin or an Iv Administration of Ciprofloxacin Dogs Parametersa (L/h/kg)b

CLENRO CLCIPRO (L/h/kg)b VssENRO (L/kg)c VssCIPRO (L/kg)c MRTENRO (h)d MRTCIPRO (h)d

1 (M)

2 (F)

3 (M)

4 (F)

5 (M)

6 (F)

7 (M)

8 (F)

Mean ± SD

0.552 0.467 2.54 2.28 4.61 4.88

0.757 0.568 2.30 1.57 3.04 2.76

0.593 0.442 2.76 1.93 4.65 4.36

0.635 0.516 2.36 2.00 3.72 3.87

0.670 0.607 2.22 2.07 3.31 3.41

0.685 0.334 1.88 1.51 2.75 4.51

1.230 0.444 3.46 2.37 2.82 5.34

0.711 0.364 2.04 1.63 2.87 4.47

0.729 ± 0.212 0.468 ± 0.094 2.45 ± 0.49 1.92 ± 0.33 3.47 ± 0.78 4.20 ± 0.82

a,b CL c ENRO and CLCIPRO, plasma clearance for enrofloxacin and ciprofloxacin, respectively (ANOVA, p < 0.05); VssENRO and VssCIPRO, steady state volume of distribution for enrofloxacin and ciprofloxacin, respectively (ANOVA, p < 0.01); dMRTENRO and MRTCIPRO, mean residence time for enrofloxacin and ciprofloxacin, respectively (ANOVA, p > 0.05). CL, Vss, and MRT were obtained from the parameters of the model (Table 1).

Table 3sMean Residence Times of Enrofloxacin and Ciprofloxacin after an Iv Administration of Enrofloxacin or Ciprofloxacin or a Po Administration of Enrofloxacin Dogs Parametersa

1 (M)

2 (F)

3 (M)

4 (F)

5 (M)

6 (F)

7 (M)

8 (F)

Mean ± SD

MRTENRO after ENROiv (h) MRTENRO after ENROpo (h) MATENRO after ENROpo (h) MRTCIPRO after ENROiv (h) MRTCIPRO after ENROpo (h) MRTCIPRO after CIPROiv (h) MFTCIPRO after ENROiv (h) MFTCIPRO after ENROpo (h)

4.61 6.37 1.76 9.56 10.70 4.88 4.68 5.82

3.04 3.72 0.68 5.85 6.07 2.76 3.09 3.31

4.65 6.84 2.19 9.06 10.71 4.36 4.70 6.35

3.72 4.44 0.72 7.65 8.03 3.87 3.78 4.16

3.31 4.12 0.80 6.80 6.98 3.41 3.28 3.46

2.75 5.80 3.05 7.32 10.04 4.51 2.81 5.53

2.82 4.03 1.21 8.20 9.16 5.34 2.85 3.82

2.87 3.83 0.96 7.42 7.59 4.47 2.95 3.12

3.47 ± 0.78 4.89 ± 1.25 1.42 ± 0.85 7.73 ± 1.20 8.66 ± 1.76 4.20 ± 0.82 3.52 ± 0.78 4.45 ± 1.26

a

MRT, mean residence time; MAT, mean absorption time; MFT, mean formation time.

(92.9 ( 2.0%), explaining the relatively great systemic bioavailability (F) of enrofloxacin (83.5 ( 7.9%) (Table 4). The fraction of the administered enrofloxacin dose metabolized to ciprofloxacin was similar after iv and po enrofloxacin administrations (40.4 ( 10.1% and 40.2 ( 8.3%, respectively) (Table 5).

Discussion This study is the first to provide a quantitative evaluation of the transformation of enrofloxacin to ciprofloxacin in an animal species. Enrofloxacin and ciprofloxacin data were analyzed with a model allowing evaluation of the fraction of enrofloxacin metabolized hepatically, the actual enrofloxacin absorption, and the first-pass hepatic loss of the drug.

Our model assumed that all ciprofloxacin was formed in the liver, this being considered as the predominant organ of quinolone metabolism.12 The parameters of the model were identifiable because enrofloxacin and ciprofloxacin plasma concentrations were measured after both iv and po enrofloxacin administrations and after iv ciprofloxacin administration. The selection of the compartmental model was based on the AIC,8 which takes into account the principle of parsimony. A general treatment for mean residence time, clearance, and volumes of distribution parameters was adopted because the existence of first-order elimination from compartments other than the sampling compartment had been evidenced.11 A socalled noncompartmental approach with computation using statistical moments can be used only if all sources and all sinks are directely associated with the probed central measurement pool(s).13 The only data set which could be analyzed

Journal of Pharmaceutical Sciences / 1153 Vol. 86, No. 10, October 1997

Table 4sAbsorption and Systemic Bioavailability of Enrofloxacin after a Po Enrofloxacin Administration Dogs Parametersa

1 (M)

2 (F)

3 (M)

4 (F)

5 (M)

6 (F)

7 (M)

8 (F)

Mean ± SD

Fabs (%) Fh (%) F (%) Eh (%)

95.56 93.55 89.40 6.45

94.20 93.39 87.97 6.61

93.46 93.82 87.68 6.18

97.53 94.44 92.11 5.56

76.66 90.15 69.10 9.85

78.99 94.64 74.75 5.36

87.38 93.49 81.69 6.51

95.29 89.33 85.12 10.67

89.88 ± 8.03 92.85 ± 1.99 83.48 ± 7.89 7.15 ± 1.99

a F , fraction of enrofloxacin absorbed into portal vein after its oral administration (see eq 2 in the text); F , fraction that survives a single passage through the liver abs h (see eq 3 in the text); F, systemic bioavailability (see eq 1 in the text); Eh, liver extraction ratio (see eq 4 in the text).

Table 5sFraction of the Administered Enrofloxacin Dose Transformed into Ciprofloxacin after Iv Enrofloxacin Administration (fm1) or a Po Enrofloxacin Administration (fm2) Dogs Parametersa

1 (M)

2 (F)

3 (M)

4 (F)

5 (M)

6 (F)

7 (M)

8 (F)

Mean ± SD

fm1 (%) fm2 (%)

36.82 39.08

35.04 37.05

42.12 42.72

44.81 46.70

39.32 34.72

34.85 30.28

62.15 56.46

28.45 34.39

40.44 ± 10.08 40.17 ± 8.33

a fm and fm , fractions of enrofloxacin metabolized to ciprofloxacin after iv and po enrofloxacin administrations, respectively. fm and fm were obtained using eq 1 2 1 2 15 and 16, respectively.

using the standard statistical moments approach was the ciprofloxacin data after iv administration, and for that data set, we obtained similar results using both compartmental or statistical moments approaches. In the present experiment, the estimated plasma clearance of enrofloxacin was 0.73 L/h/kg, a value lower than that reported by others (1.62 L/h/kg)3 and (1.26 L/h/kg).14 The origin of such a difference remains unclear. However, it should be noted that a high plasma enrofloxacin clearance is not consistent with the hypothesis of an enrofloxacin transformation in the liver when an almost total oral bioavailability is measured.3 The possibility of loss of enrofloxacin during storage leading to the hypothesis that a fraction of the enrofloxacin might be converted to ciprofloxacin during storage, and thus lead to an overestimation of enrofloxacin clearance was unlikely. In contrast to Kaartinen et al.,2 we have demonstrated that enrofloxacin concentrations were stable under our storage conditions for at least 2 years; moreover, our plasma samples were analyzed rapidly after completion of the experiment (within 4 months instead of the 18 months cited by the aforementioned investigators). After an iv enrofloxacin administration, we found that 40% of the enrofloxacin was metabolized to ciprofloxacin, a more potent quinolone than enrofloxacin, thus supporting the claim that a significant proportion of the enrofloxacin activity was actually due to ciprofloxacin.3 Enrofloxacin is a drug with a low hepatic extraction ratio (Eh ) 11%) with Eh equal to the ENRO , i.e. about 0.3 ratio of hepatic clearance (here fm1CLENRO,iv L/h/kg) to the hepatic blood flow15 (i.e. about 2.52 L/h/kg in dog16). Thus, the hepatic clearance of enrofloxacin should be determined by only two physiological factors: its plasma protein binding and its intrinsic clearance, i.e. the level of the hepatic enzyme activity. The protein binding of enrofloxacin in dog is low (15-20 %).17 It can therefore be assumed that the hepatic clearance of enrofloxacin, i.e. transformation of enrofloxacin to the potent ciprofloxacin, will only depend on liver enzymatic activity. The relatively low hepatic extraction ratio of enrofloxacin explains its rather good systemic availability after an oral route of administration. In the present experiment, the estimated systemic enrofloxacin bioavailability was 83%, a limited fraction of enrofloxacin (about 10%) not being absorbed and another small fraction (about 7% of the absorbed fraction) undergoing a liver first-pass metabolism. This observed extraction ratio is consistent with what we estimated from the iv study i.e. Eh ) 11% (see above). This means that enrofloxacin bioavailability can in practice be 1154 / Journal of Pharmaceutical Sciences Vol. 86, No. 10, October 1997

influenced not only by the process of absorption per se but also by enzymatic liver activity, an increase in hepatic metabolism leading to a more extensive presystemic transformation of enrofloxacin to ciprofloxacin. This presystemic transformation was limited under our conditions and iv and po routes could be considered similar from a therapeutic point of view, the incomplete absorption of enrofloxacin being compensated by the first-pass hepatic formation of the more potent ciprofloxacin. The mean time of ciprofloxacin formation from enrofloxacin was relatively long (about 3.52 and 4.46 h after iv and po enrofloxacin administrations, respectively). This is explained by both the low hepatic enrofloxacin clearance and the large enrofloxacin volume of distribution (Vss ) 2.45 L/kg). Such a large distribution is due to the rather lipophilic nature of fluoroquinolones and their low degree of ionization. In the present experiment, the estimated plasma clearance of ciprofloxacin was 0.47 L/h/kg, a value significantly lower than that reported by others in dog (about 1 L/h/kg).18 The plasma clearance of ciprofloxacin was significantly lower than that of enrofloxacin (0.73 L/h/kg) and the Vss of ciprofloxacin was significantly lower than that of enrofloxacin (1.92 versus 2.45 L/kg), explaining a longer but not signicantly different mean residence time for ciprofloxacin than for enrofloxacin (4.20 h versus 3.47 h). It can be concluded that the six-compartment model herein described has allowed the investigation of both enrofloxacin and ciprofloxacin disposition and the extent of enrofloxacin to ciprofloxacin transformation after iv and po administrations. It has been shown that enrofloxacin is largely metabolized to ciprofloxacin in dog and that the fractions of metabolized enrofloxacin were similar after intravenous and oral administrations of enrofloxacin, the hepatic first-pass effect being low.

References and Notes 1. Pyo¨ra¨la¨, S.; Panu, S.; Kaartinen, L. In Proceedings of the 6th International Congress of the European Association for Veterinary Pharmacology and Toxicology, Edinburgh, UK, August 7-11, 1994. Lees, P., Ed.; Blackwell Scientific Publications: Oxford, 1994; pp 45. 2. Kaartinen, L.; Salonen, M.; A ¨ lli, L.; Pyo¨ra¨la¨, S. J. Vet. Pharmacol. Ther. 1995, 18, 357-362. 3. Ku¨ng, K.; Riond, J. L.; Wanner, M. J. Vet. Pharmacol. Ther. 1993, 16, 462-468.

4. Prescott, J. F.; Yelding, K. M. Can. J. Vet. Res. 1990, 54, 195197. 5. Mevius, D. J.; Breukink, H. J.; Van Miert, A. S. J. P. A. M. Vet. Quat. 1990, 12, 212-220. 6. Shah, V. P.; Midha, K. K.; Dighe, S; McGilveray, I. J.; Skelly, J. P.; Yacobi, A.; Layloff, T.; Viswanathan, C. T.; Cook, C. E.; McDowall, R. D.; Pittman, K. A.; Spector, S. J. Pharm. Sci. 1992, 81, 309-312. 7. Schneider, M.; Thomas, V.; Boisrame´, B.; Deleforge, J. J. Vet. Pharmacol. Ther. 1996, 19, 56-61. 8. Yamaoka, K. ; Nakagawa, T. ; Uno, T. J. Pharmacokinet. Biopharm. 1978, 6, 165-175. 9. Gibaldi, M.; Perrier, D. Pharmacokinetics; Swarbrick, J., ed.; Marcel Dekker: New York, 1982. 10. Koeppe, P.; Hamann, C. Comput. Program Biomed. 1980, 12, 121-128.

11. Nakashima, E.; Benet, L. Z. J. Pharmacokin. Biopharm. 1988, 16, 475-492. 12. So¨rgel, F.; Jaehde, U.; Naber, K.; Stephan, U. Clin. Pharmacokinet. 1989, 16, 5-24. 13. DiStefano, J. J. Am. J. Physiol. 1982, 243 (1), R1-R6. 14. Intorre, L.; Mengozzi, G.; Maccheroni, M.; Bertini, S.; Soldani, G. J. Vet. Pharmacol. Ther. 1995, 18, 352-356. 15. Rowland, M.; Tozer, T. N. Clinical pharmacokinetics: Concepts and applications, Lea & Febiger: Philadelphia, 1989. 16. Boxenbaum, H. J. Pharmacokinet. Biopharm. 1980, 8, 165-176. 17. Petzinger, E. Tiera¨ rztliche Praxis 1991, 19, 14-20. 18. Abadia, A. R.; Aramayona, J. J.; Mun˜oz, M. J.; Pla Delfina, J. M.; Saez, M. P.; Bregante, M. A. J. Vet. Pharmacol. Ther. 1994, 17, 384-388.

JS9603461

Journal of Pharmaceutical Sciences / 1155 Vol. 86, No. 10, October 1997