plasma concentrations resulting from florfenicol

PLASMA CONCENTRATIONS RESULTING FROM FLORFENICOL PREPARATIONS GIVEN TO PIGS IN THEIR DRINKING WATER L. Gutiérrez, D. Vargas, L. Ocampo, H. Sumano, R. Martinez and G. Tapia J ANIM SCI published online March 31, 2011

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://www.journalofanimalscience.org/content/early/2011/03/31/jas.2010-3576

www.asas.org

Downloaded from www.journalofanimalscience.org by guest on March 15, 2013

Plasma concentrations of florfenicol in pigs PLASMA CONCENTRATIONS RESULTING FROM FLORFENICOL PREPARATIONS GIVEN TO PIGS IN THEIR DRINKING WATER1 L. Gutiérrez*, D. Vargas*, L. Ocampo*, H. Sumano2*, R. Martinez, G. Tapia¶ *Departamento de Fisiología y Farmacología, Facultad de Medicina Veterinaria y Zootecnia. Universidad Nacional Autónoma de México. Av. Universidad 3000, Coyoacán, México City 04510  Centro de Enseñanza, Investigación y Extensión en Producción Porcina, Facultad de Medicina Veterinaria y Zootecnia. Universidad Nacional Autónoma de México Km. 2 de la carretera Jilotepec-Corrales, Jilotepec, Estado de México. ¶

Departamento de Genética y Bioestadística, Facultad de Medicina Veterinaria y Zootecnia, Universidad Nacional Autónoma de México. Av. Universidad 3000, Coyoacán, México City 04510

1

Acknowledgements: Our study was supported by a grant (60303) from Secretaria de Educacion Publica-Consejo Nacional de Ciencia y Tecnologia (SEP-Conacyt), Mexico. 2 Corresponding author: Phone: (52) 55 5 622 5908. email: [email protected]

1 Downloaded www.journalofanimalscience.org guest on March 15, 2013 Published Onlinefrom First on March 31, 2011 asbydoi:10.2527/jas.2010-3576

ABSTRACT Florfenicol administered through the drinking water has been recommended as a metaphylactic antibacterial drug to control outbreaks of respiratory diseases in pigs caused by strains of Actinobacillus pleuropneumoniae and Pasteurella multocida. Yet, it is difficult to pin point in practice when the drug is given metaphylactically or therapeutically. Further, pigs are likely to reject florfenicol-medicated water, and plasma concentrations of the drug are likely to be marginal for diseases caused by Escherichia coli, Klebsiella pneumoniae, and Staphylococcus aureus. The reported minimal inhibitory concentration (MIC) values for these organisms show a breakpoint of 2 to 3 μg·mL-1. An experiment was conducted during September and October 2009. One hundred twenty healthy crossbred pigs (Landrace/Yorkshire), weighing 23 ± 6.2 kg, were used in this trial. They were randomly assigned to 5 groups with 3 replicates of 8 animals per group. Two commercial preparations of florfenicol were administered through drinking water at 2 concentrations (0.01% and 0.015%). Water intake was measured before and after medication, and plasma concentrations of florfenicol were determined by HPLC. Considerable rejection of florfenicol-medicated water was observed. However, plasma florfenicol concentrations were of a range sufficient for a methaphylaxis approach to preventing disease by bacteria with MIC breakpoints  0.25 μg·mL-1. However, lower efficacy as a metaphylactic medication should be expected for bacteria with MIC > 0.25 μg•mL-1, considering the reported existence of resistant bacteria to florfenicol and natural resistance of Streptococcus suis or E. coli to this drug.

Key Words: Florfenicol, metaphylaxis, pigs, plasma-concentrations, water-medication.

2 Downloaded from www.journalofanimalscience.org by guest on March 15, 2013

INTRODUCTION Florfenicol has been recommended as metaphylactic antibacterial drug for acute outbreaks of respiratory diseases caused by Actinobacillus pleuropneumoniae and Pasteurella multocida. An average minimal inhibitory concentration (MIC) of 0.25 µg·mL-1 was reported for 108 A. pleuropneumoniae strains isolated from pig lungs in Italy (Barigazzi et al., 1996). For Escherichia coli, Klebsiella pneumoniae and Staphylococcus aureus the reported MICs resulted in larger values (> 3.1 μg·mL-1) (Neu and Fu, 1980). The recommended dose of the commercially available preparation for intramuscular (IM) use is 15 to 20 mg•kg-1 BW every 48 h. In contrast, in drinking water, a concentration of 106 μg•mL-1 is recommended for 5 consecutive days. If water consumption is 120 mL·kg-1/d (Froese, 2003) then, an approximate dose of 10 – 12 mg•kg-1 is theoretically, being delivered. Manufacturers specify that medication should be initiated promptly when a swine respiratory disease is predictable. This is possibly due to the fact that an approximately a 50% reduction in water intake reported for pigs infected with Actinobacillus pleuropneumoniae (Swinkels et al., 1994).

Although it is known that plasma concentrations of the drug must be elevated above 0.25 µg·mL1

to achieve metaphylaxis in pigs, it is not known how group housing with drug delivery through

water will on average affect plasma concentrations of the drug. Values derived from pharmacokinetic (PK) studies do not necessarily reflect serum concentrations of the same drug obtained under unaltered field conditions in a productive unit. In the latter setting, water is available ad libitum. In contrast, in PK studies the drug is administered as a single oral gavage (Liu et al., 2003). Also, considering that patent protection for the reference preparation is still


valid, vehicles used in a given generic formulation are likely to be different, and in turn this could make acceptance of medicated water by pigs also different.

There are no specific data of the drinking habits of healthy pigs confronted with a single source of water containing florfenicol. Hence, this study was conducted to assess: (1) the drinking patterns of healthy pigs when changing from potable non-medicated water to water carrying concentrations of florfenicol of either 0.01% (100 μg•mL-1) or 0.015% (150 μg•mL-1), and (2) the plasma concentrations of florfenicol achieved in this species under such conditions, using 2 commercially available formulations: the reference and the generic one.

MATERIAL AND METHODS The study was given ethical approval by the Internal Committee of Postgraduate Studies of the Faculty of Veterinary Medicine and Animal Husbandry of the Universidad Nacional Autonoma de Mexico. It was carried out at the experimental farm of the same institution, located in Jilotepec, State of Mexico, throughout the temperate months of September and October 2009, with a mean environmental temperature of 16 ± 8.8° C and a mean inside-pen temperature of 19.5 ± 4.2° C. In all, 120 healthy crossbred, castrated male pigs (Landrace/Yorkshire), weighing 23 ± 6.2 kg were included in this trial. Negative tests for A. pleuropneumoniae were determined by slide agglutination against serotype 1 (Pleurotest, Ciprolab, Mexico) during a 2-wk quarantine period. Pigs were fed a commercial feed free of drugs for ad libitum intake (La Hacienda S.A de C.V., Toluca, México) having CP 16.0%, fat 3.0%, crude fiber 6.5 %, humidity 1.0 %, ashes 7.0%, N-free elements 55.5%.


The experimental design was complete randomized with treatments arranged in a 2×2 factorial array with an additional control. The first primary factor was the product with 2 levels: florfenicol (Nuflor 10% oral solution, Intervet/Schering Animal Health, Mexico City, Mexico) and a generic preparation of florfenicol (10% solution, Laboratorios Aranda, Querétaro Qro., Mexico). Second primary factor was the dose with 2 levels: 0.01% and 0.015%. A control group without medication was added to compare the water consumptions as a baseline. Based on this, treatments were arranged as a 2×2+1. The 5 groups resulting: reference florfenicol at 0.01% (FLOR0.01), reference florfenicol at 0.015% (FLOR0.015), generic florfenicol at 0.01% (GEN0.01), generic florfenicol at 0.015% (GEN0.015) and non medicated control group (CONT). All groups were replicated 3 times. Treatments were randomly assigned by means of a table of random numbers to 120 pigs, restricted to allocate exactly 8 pigs in each replicate for each treatment. Every group of eight pigs, housed in 4-m2 pens, was considered an experimental unit.

Water was supplied by placing 30-L graduated containers hanging from each pen with drinking nipples (Trojan Model 33CI , Vittetoe, Inc, IA, USA) set at 38 cm from the floor (2 per pen). During the 2 wk previous to this trial, water disappearance as an assessment of water intake was measured daily and animals weighed weekly. Water intake was measured and reported as mL/group. At the beginning of the 3 wk, the water source was medicated as described (0.01% or 0.015%), and offered to pigs as the only source of water for 3 d, starting every day at 0700-h and withdrawing it at 2200-h. Access to water was denied overnight (9-h). This is not a common practice, but previous experiences demonstrated that pigs tend to play with the water supply altering water intake measurement. Aided by 6 trained technicians blinded to treatment,


medicated-water intake was monitored hourly and final volumes recorded at 2200-h. During the day, water containers were refilled as needed with medicated water. Although this trial was projected to a 5-d period of water medication, pigs included in this trial were healthy subjects, and on d 3 an almost complete refusal to drink medicated water raised the issue of animal welfare. Hence, it was decided that the medication be applied for only 3 d.

During the 3-d period of medication, blood samples were obtained by direct jugular puncture from 6 pigs per group at each time, thus bleeding randomly each pig only 3 to 4 times during the day. Then, 3 repetitions were obtained. Daily sampling times per day were: 0800 h, 1000 h, 1200 h, 1400 h ,1600 h, 2000 h and at 2400 h on the third day. Five-milliliter syringes, previously charged with 10 IU of heparin, were used. To differentiate pigs that were bled from the ones that needed to be bled, different colors of paint applied to the back were used each day. Blood was immediately centrifuged at 3,000 × g for 10 min at room temperature, plasma separated in individual Eppendorf containers, labeled and then stored-frozen in liquid nitrogen.

An HPLC method was established in this study for the measurement of florfenicol concentrations in pig plasma, based on the technique described by Liu et al. (2003), using chloramphenicol as internal standard and mobile phase of 25% acetonitrile in water with a flow of 1.5 mL·min-1. On the day of analysis, the thawed plasma samples were spiked with the internal standard and extracted with ethyl acetate by agitation and centrifugation. After evaporation of the organic material under N flow, the residue was reconstituted in the mobile phase and injected onto a Jasco XLC liquid chromatograph (Evoelution S.A. de C.V. Mexico City) consisting of a variable-wavelength detector UV set at 224 nm. A reversed-phase C18 column (4 µm, 4.6 x 250


mm; C18, Alltech Associates Inc., Deerfield IL) was fitted and eluted with aqueous acetonitrile (77:23, vol·vol-1) at a flow rate of 1 mL•min-1 at 25°C. A linear relationship existed in the calibration curve over the range of 0.05 to 20 μg•mL-1 (r > 0.995). The within-run and betweenrun precision and recovery percentage for florfenicol were: ≤ 5, ≤ 6, and 92%, respectively. Limit of quantification was 0.025 µg·mL-1, and limit of detection 0.005 µg·mL-1.

Area under the plasma concentration (AUC) versus time curves were obtained through the trapezoidal method and the ratios of AUC·MIC-1 with 2 MIC values were calculated for each group. Area under the plasma concentration values were compared between groups by means of ANOVA and Bonferroni t tests. Statistical analysis of water consumption before and after medication was carried out by means of the repeated measure model with treatments arranged as 2×2+1 array (General Linear Model; Cox et al., 1999) with the statistical package SPSS 16 (SPSS Inc., Chicago, IL) with variable products and dose as between-subject variables and sampling as within-subject factor, considering each pen as a subject. Greenhouse-Geisser epsilon was used for correction to significant Mauchly’s test of sphericity. The control group was added to provide a comparative baseline. Dunnett post hoc multiple comparison tests were done for significant variables. Plasma concentrations of florfenicol in pigs were transformed to Y = log (x) to carry out the Gaussian model assumption, and data was analyzed by the repeated measure model as previously outlined. For all tests significance level (α) was set as 0.05. Control group was analyzed to ensure that no florfenicol was present. The G*Power program (http://www.psycho.uni-duesseldorf.de/aap/projects/gpower/) was used to determine the sample size to take from each treatment group (Faul et al., 2007).


RESULTS Water consumption levels before (on weekly basis) and after (on daily basis) medication are presented in Figure 1. A sharp decrease of approximately 30% to 34% in water intake was observed in all groups (P = 0.01) on the first day of medication. This trend was again observed on the second day of medication with a 55% to 62% reduction in water intake as compared to the mean basal values. On the third day, reduction in water consumption reached a maximum of 65% to 68% from basal values. Reduction in water consumption on this day and the day before were statistically significant both from basal values (P = 0.01), and from day to day (F = 21.143, df = 11.11; Greenhouse-Geisser correction = 27.77, ε = 0.694; P > 0.001). The CONT group showed no variation in this parameter, but was statistically different from treated groups (P = 0.01).

Control group was found free of florfenicol. Plasma concentrations of florfenicol in all groups are presented in Figure 2. In the CONT group all plasma samples were negative to florfenicol as quantified by the referred technique. No differences were shown for the product * hour interaction (F = 1.016, df = 25.55; Greenhouse-Geisser correction = 170.32, ε = 0.532; P = 0.510). The factor product shows statistical significance (F = 3.55, df = 3, 30; P = 0.033) in the tests between-subjects effects of the repeated model analysis. This means that dose level has an effect in the plasma concentrations of the drug. In the pairwise comparisons, the factor dose


showed a lower plasma concentration of florfenicol (P = 0.005) for the generic preparation at 0.010%, as compared to the reference preparation at 0.015%. But when the comparison was done between reference and generic preparations at the same concentrations (0.010% and 0.015%) there were no differences (P = 0.415 and 0.120, respectively).

Statistical comparisons of AUC values and AUC·MIC-1 ratios are presented in Table 1. This table also shows calculated dose of florfenicol based on real mean water consumption observed. Two AUC·MIC-1 ratios are obtained and listed, considering 2 breakpoints: one representing sensitive bacteria such as A. pleuropneumoniae (0.25 μg•mL-1) and the other representing more resistant bacteria, such as E. coli (2.0 μg•mL-1). Plasma concentrations of florfenicol versus time are presented in Figure 2, pointing out the referred breakpoints.

DISCUSSION

The rapid slide agglutination test used in this trial does not have the same sensitivity as that obtained with ELISA or polymerase chain reaction to fully identify serotypes of A. pleuropneumoniae. However, it is both quick and simple and has become a routine method for initial diagnosis of A. pleuropneumoniae field outbreaks (Mittal et al., 1987; Rapp et al., 1985). Negative results from this test revealed the lack of an outbreak of swine pleuropneumonia at the beginning of the medication with florfenicol-in-water, and therefore, patterns of florfenicolmedicated water-intake observed in this trial should not be linked to an initial outbreak of the disease. It is also worth emphasizing that in our experience blood sampling schedule is unlikely to negatively influence water consumption in healthy pigs; yet in future studies, a positive


control pen receiving florfenicol through water and not being bled should be studied to test the influence of stress caused by repeated bleedings on water consumption.

As observed by Liu et al. (2003), the HPLC analytical technique used, show 92% recovery in florfenicol-enriched samples, as well as relatively small intra-assay and inter-assay errors. Thus, it is unlikely that a large quantity of florfenicol could have escaped quantification in plasma samples. Hence, it appears reasonable to assume that these results are quantitatively reliable and indicate that pigs receiving florfenicol as in-water medication (as 0.01% or 0.015%) have effective plasma concentrations of this drug for most A. pleuropneumoniae strains.

Water consumption showed a sharp reduction, from basal values on the first day. On the second and third day, pigs medicated with florfenicol at a rate of 0.01% showed a further decrease in water intake, to reach a minimum, almost 25% of the total water intake registered in the CONT group. Groups receiving florfenicol at a rate of 0.015% had an even more pronounced reduction in water intake. Initial dose offered to pigs was 9 or 13.5 mg•kg-1, whether in the groups medicated at a rate of 0.01% or at a rate of 0.015%, respectively. The reduction in water intake on d 3 was so severe that it could have affected the animals’ welfare. Additionally, such reduction trimmed down calculated dose to 4.2 to 5.5 mg•kg-1 for the 0.01% and 0.015% groups, respectively. Consequently, plasma concentrations of florfenicol in this study indicate an apparent lower bioavailability as compared to studies of pharmacokinetics detailed by Liu et al. (2003). This may be explained by a smaller dose delivered to pigs, derived from the lower water intake observed in this study plus the non-quantified water spillage.


Also, significant lower concentrations in plasma (P = 0.027) for the generic preparation of florfenicol in 0.010% concentration were found, but within the accepted limits to be regarded as bioequivalent. It also appears that reduction in water intake occurs independently of the preparation of florfenicol used. Because Nuflor vehicles are patent-protected, the corresponding vehicles in the 2 florfenicol preparations cannot be the same; therefore, florfenicol is the common ingredient and likely cause of the observed reduction in water intake.

AliAbsdi and Lees (2000) suggested that the dosage regimen for the treatment of animals should be optimized, hence maintaining concentrations at the site of infection in excess of MIC90 for the entire medication period for bacteriostatic drugs such as florfenicol, a drug acting primarily by time-dependant mechanisms. The purpose of determining the MIC is to attempt to integrate the potential efficacy of the target population of pathogens with the pharmacokinetics of the antibiotic. Breakpoints are discriminatory concentrations based on pharmacokinetic data in the relevant animal species and are used to interpret MIC values as susceptible. For example, if an average MIC of 0.25 µg·mL-1 is reported as the breakpoint for A. pleuropneumoniae (Barigazzi et al., 1996), then all groups medicated with florfenicol exhibit plasma concentrations of florfenicol that surpass this value throughout the dosing interval. If a greater MIC value is considered i.e., 3.1 µg·mL-1 for Streptococcus suis or E. coli (Priebe and Schwarz, 2003; Ueda and Suenaga, 1995), florfenicol as in water medication would not be recommended. Furthermore, this trial was conducted with clinically healthy animals, but it is likely that a more pronounced reduction in water intake will occur in pigs suffering clinical disease whose water intake has been calculated to be reduced by half their normal water intake before medication (Swinkels et al., 1994).


In this study, AUC values obtained by the trapezoidal method, ranged from 28 ± 4 and 30 ± 4 µg·h-1·mL-1 on d 1 to 22 ± 3 and 23 ± 2 µg·h-1·mL-1 on d 3 of medication in groups 0.01% and 0.15%, respectively. In studies with florfenicol administered orally as a bolus dose by gavage after a 16-h fast, the AUC values were 66 and 73 µg·h-1·mL-1, whether derived from healthy or pneumonic pigs (Liu et al., 2003). With a breakpoint of 0.25 µg·mL-1, and considering AUC·MIC-1 as a reliable predictor of clinical efficacy (AliAbsdi and Lees, 2000), AUC·MIC-1 ratios will range in this study from approximately 120 to 290, taking values found by Liu et al. (2003). With a more challenging bacterial resistance scenario (i.e., 2 µg·mL-1), AUC·MIC-1 ratio will reach a maximum value of 15 in the present study. Under field conditions, this latter value is likely to result in an evident reduction in cure-rate. Nevertheless, it is worth emphasizing that optimal AUC·MIC-1 ratios have not yet been established for florfenicol. However, plasma concentrations found in this trial can be regarded as suboptimal for the treatment of Streptococcus suis, E. coli or resistant Bordetella bronchiseptica related diseases (Kadlec et al., 2007; Kehrenberg et al., 2004; Priebe and Schwarz, 2003).

In conclusion, plasma concentrations of florfenicol in pigs derived from dosing under almost unaltered field surroundings are very different from PK studies carried out under control conditions, i.e., the drug is administered not as a single bolus dose but through medicated water, available to the animal for ad libitum intake. This study shows that florfenicol can only be used as a metaphylactic antibacterial strategy and only against a limited number of bacteria of the “porcine respiratory disease complex”. Reduction in water intake in sick animals impedes its use


as a therapeutic alternative. Modification of taste in florfenicol preparations for pigs is postulated as an important line of research for pharmaceutical designers.

LITERATURE CITED. AliAbsdi, F. S., and P. Lees. 2000. Antibiotic treatment for animals: effect on bacterial population and dosage regimen optimization. Int. J. Antimicrob. Agents. 14:307–313. Barigazzi, G. P., C. Candott, and E. Foni. 1996. In vitro susceptibility of 108 isolated Actinobacillus pleuropneumoniae strains to 17 antimicrobial agents from pig lungs in Italy. Page 207 in Proc. 14th Int. Pig Vet. Soc. Congr. Bologna, Italy. Cox, E. H., C. Veyrat-Follet, S. L. Beal, E. Fuseau, S. Kenkare, and L. B. Sheiner. 1999. A population pharmacokinetic-pharmacodynamic analysis of repeated measures time-to-event pharmacodynamic responses: the antiemetic effect of ondansetron. J. Pharmacokinet. Biopharm. 27: 625-44. Faul, F., E. Erdfelder, A. G. Lang,. and A. Buchner. 2007. G*power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavoir Res. Methods. 39:175-191. Froese, C., 2003. Water usage and manure production rates in today´s pig industry. Banff Pork Seminar Proceedings, Advances in Pork production. 16:215-223.


Kadlec, K., C. Kehrenberg, and S. Schwarz. 2007. Efflux-mediated resistance to florfenicol and/or chloramphenicol in Bordetella bronchiseptica: identification of a novel chloramphenicol exporter. J. Antimicrob. Chemother. 59:191–196. Kehrenberg, C., J. Mumme, J. Wallmann, J. Versphol, R. Tegeler, T. Kühn, and S. Schwarz S. 2004. Monitoring of florfenicol susceptibility among bovine and porcine respiratory tract pathogens collected in Germany during the years 2002 and 2003. J. Antimicrob. Chemother. 54:572–574. Liu, J., K. F. Fung, Z. Chen, Z. Zeng, and J. Zhang. 2003. Pharmacokinetics of Florfenicol in Healthy Pigs and in Pigs Experimentally Infected with Actinobacillus pleuropneumoniae. Antimicrob. Agents. Chemother. 47:820–823. Mittal, K. R., R. Higgins, and S. Lariviere. 1987. An evaluation of agglutination and coagglutination techniques for serotyping of Haemophilus pleuropneumoniae isolates. Am. J. Vet. Res. 48:219-226. Neu, K. P. and H. C Fu. 1980. In vitro activity of chloramphenicol and thiamphenicol analogs. Antimicrob. Agents. Chemother. 18: 311–316. Priebe, S. and S. Schwarz. 2003 In vitro activities of florfenicol against bovine and porcine respiratory tract pathogens. Antimicrob. Agents. Chemother. 47:2703–2705. Rapp, V. J., R. F. Ross, and B. Z. Erickson. 1985. Serotyping of Haemophilus pleuropneumoniae by rapid slide agglutination and indirect fluorescent antibody tests in swine. Am. J. Vet. Res. 46:185-192. Swinkels, J. M., A. Pijpers, S. J. C. M. Vernooy, A. Van-Nes, and J. H. M. Verheijden. 1994. Effects of ketoprofen and flunixin in pigs experimentally infected with Actinobacillus pleuropneumoniae. J. Vet. Pharmacol. Therap. 17:299–303.


Ueda, Y. and I. Suenaga. 1995.

In vitro antibacterial activity of florfenicol against

Actinobacillus pleuropneumoniae. J. Vet. Med. Sci. 57:363-364.


Table 1. Area under the curve (AUC) for plasma concentration of florfenicol vs. time , which was obtained through the trapezoidal method and the ratios of AUC·MIC-1 with 2 MIC values (0.25 µg·mL-1 and 2.0 µg·mL-1). Values were calculated for 4 groups with 3 replicates of 8 pigs per group, receiving either 0.01% or 0.015% of the reference or a generic preparation of florfenicol in drinking water. Reference preparation Generic preparation Day FLOR0.01 FLOR0.015 GEN0.01 GEN0.015 X ± SE

1

2

3

a, b, c y,z

AUC AUC·MIC-1 0.25 µg·mL-1 AUC·MIC-1 2.0 µg·mL-1 AUC AUC·MIC-1 0.25 µg·mL-1 AUC·MIC-1 2.0 µg·mL-1 AUC AUC·MIC-1 0.25 µg·mL-1 AUC·MIC-1 2.0 µg·mL-1

28.31±4.21a,c,y 113.24 14.16 25.67±3.15a,b,y,z 102.68

X ± SE

29.77±4.13c,x 119.08 14.89 27.64±3.09b,x 110.56

X ± SE

23.63±4.23b,x 94.52

X ± SE

25.31±4.18a,b,x 101.24

11.81 23.06±3.12a,x 92.24

12.65 24.97±3.08a,b,x 99.88

12.83

13.82

11.53

12.48

21.63±2.65a,z 86.52

22.85±2.43a,z 91.40

20.09±2.36a,x 80.36

21.44±2.32a,x 85.76

10.81

11.42

10.04

10.574

Different letter indicates statistically significant difference in rows (P < 0.05) Different letter indicates statistically significant difference in columns (P < 0.05).


Group FLOR0.01 Group B FLOR0.015 Group C GEN0.01 Group D GEN0.015 Group CONT

35

30

20

-1

L·pen ·day

-1

25

15

10

5

0 -14 -14

-7 -7

00 Days

1 1

22

33

Fig. 1. Mean (± SE) water intake (L per pen per day) in 5 groups with 3 replicates of 8 pigs per group, before in-water medication (on weekly basis: -14 and -7 d) or after medication on daily basis (1, 2, 3 d) with florfenicol at 2 dose rates (0.01 and 0.015%) with either the reference (FLOR0.01, FLOR0.015) or a generic preparation (GEN0.01 and GEN0.015). A sharp decrease of approximately 30% to 34% in water intake was observed in all groups (P = 0.01), on the first day of medication, as compared with the control group (CONT).


Manuscript ID E-2010-3576.R1 Figure 1

2.2 Breakpoint = 2.0

2.0

Group GEN0.015 Group GEN0.01 Group FLOR0.015 Group FLOR0.01

Concentrations in µg·mL

-1

1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2

Breakpoint < 0.25

0.0 0

4

8

12

16

20

24

28

32

36

40

44

48

52

56

60

64

68

72

Time in hours

Fig. 2. Mean (± 1 SE) plasma concentrations of florfenicol in pigs receiving either 0.01% or 0.015% of the reference (FLOR0.01 and FLOR0.015) or a generic preparation of the drug (GEN0.01 and GEN0.015) administered through their drinking water, in 4 groups with 3 replicates of 8 pigs per group. Statistical analysis showed that interaction of time vs. group is not statistically significant (P = 0.510); however, all groups showed a decrease of drug in plasma concentrations (P < 0.05). Two breakpoints are considered: one representing sensitive bacteria such as A. pleuropneumoniae (0.25 μg·mL-1) and the other representing more resistant bacteria, such as E. coli (2.0 μg·mL-1).
