S

A(

DJa

b

c

a

ARRA

KEDWR

1

cbiit(tc2s(Em

0d

Preventive Veterinary Medicine 105 (2012) 160– 163

Contents lists available at SciVerse ScienceDirect

Preventive Veterinary Medicine

j our na l ho me p age: ww w.elsev ier .com/ locate /prevetmed

hort communication

ccumulation and depletion kinetics of erythromycin in rainbow troutOncorhynchus mykiss)

aniel Vendrell a, Lidia Serarolsb, José Luis Balcázarc,∗, Ignacio de Blasa, Olivia Gironésa,osé Luis Múzquiza, Imanol Ruiz-Zarzuelaa

Laboratory of Fish Pathology, Universidad de Zaragoza, c/Miguel Servet 177, 50013 Zaragoza, SpainLaboratorios HIPRA S.A., Avda. La Selva 135, 17170 Amer (Girona), SpainCatalan Institute for Water Research (ICRA), Parc Científic i Tecnològic de la Universitat de Girona, 17003 Girona, Spain

r t i c l e i n f o

rticle history:eceived 13 October 2011eceived in revised form 1 February 2012ccepted 21 February 2012

eywords:

a b s t r a c t

Erythromycin is an antimicrobial agent recommended for the control and treatment ofdiseases caused by Gram-positive bacteria. Few studies, however, have determined themetabolic and pharmacokinetic aspects of this antimicrobial agent in fish. The aim of thepresent study, therefore, was to determine the accumulation and depletion time of ery-thromycin after administration of medicated feed containing 52 mg kg−1 body weight day−1

for 8 days in rainbow trout (Oncorhynchus mykiss). Results were analyzed following the

rythromycinepletionithdrawal time

ainbow trout

European Agency for Evaluation of Medicinal Products guidelines. We measured a with-drawal time of 187 ◦C-day (◦C-day = water temperature × days), lower than the value(500 ◦C-day) recommended by Council Directive 2004/28/EC for veterinary medicinal prod-ucts. Our results provide data to establish therapeutic regimens for the use of erythromycinin aquaculture.

. Introduction

World aquaculture production has increased signifi-antly in the last few years; however, its development haseen accompanied by the spread of pathogens and by an

ncrease in the demand for veterinary drugs. Erythromycins an antimicrobial agent recommended for the control andreatment of diseases caused by Gram-positive bacteriaMunday, 1994). Few studies, however, have determinedhe metabolic and pharmacokinetic aspects of this antimi-robial agent in fish (Fairgrieve et al., 2006; Esposito et al.,007). The aim of the present study, therefore, was totudy plasma levels of erythromycin in rainbow trout

Oncorhynchus mykiss) during treatment with Hipramix-rythromycin (Laboratorios HIPRA S.A., Girona, Spain) viaedicated feed to determine whether the proposed dosage

∗ Corresponding author. Tel.: +34 976761569; fax: +34 976761612.E-mail address: jlbalcazar@icra.cat (J.L. Balcázar).

167-5877/$ – see front matter © 2012 Elsevier B.V. All rights reserved.oi:10.1016/j.prevetmed.2012.02.017

© 2012 Elsevier B.V. All rights reserved.

is suitable to develop and maintain therapeutic levelsagainst the pathogenic bacteria Lactococcus garvieae. Wealso studied the depletion of erythromycin residues in fishtissues to establish the length of the necessary withdrawalperiod before sacrificing the fish for intended human con-sumption.

2. Materials and methods

2.1. Fish and experimental conditions

Two hundred healthy rainbow trout with aver-age weight 152 ± 4 g (mean ± standard deviation)were obtained from a commercial fish farm from theAutonomous Community of Aragon, Spain. The health sta-

tus was examined using conventional microbiological testsand PCR as previously described (Vendrell et al., 2007). Thefish were then randomly divided into four 1000-l tanks,each containing 50 fish. After the acclimatization period,

terinary

D. Vendrell et al. / Preventive Ve

all fish were maintained with a continuous flow of aeratedfresh water at 18 ± 1 ◦C.

2.2. Antibiotic administration and sample collection

We used a commercial fish feed (Skretting, Bur-gos, Spain) supplemented with Hipramix-Erythromycin(14 kg t−1). The final erythromycin (erythromycin thio-cyanate) concentration in the feed was 2600 ppm. All fishwere fed the medicated diet once a day for 8 consecu-tive days at 2% body weight day−1, after which fish wereswitched to an unmedicated diet until the end of the trial.Feeding rate was adjusted after each sampling based onthe mean weight of the fish sampled. The dosage used inthe medicated diet was selected according to the feedingrate and therapeutic doses suggested for fish disease treat-ment (about 52 mg erythromycin kg−1 body weight day−1).The tanks were cleaned daily to avoid fish re-ingestingexcretions containing antibiotic residues and possible left-over medicated feed.

Fish were randomly sampled from each experimentaltank (n = 10 fish for each time interval) in order to obtainplasma samples during the 8 days of treatment and on days2, 4 and 6 post-treatment. With regard to the study of tissuedepletion of erythromycin, groups of 10 fish were sacrificedand tissue samples were collected on days 2, 4, 6, 8, 10, 15,20 and 28 after the treatment. Blood samples (1.0 ml) werewithdrawn from the caudal vein using sterile syringes andplaced in heparinized tubes. The plasma was obtained fromthe blood samples by centrifugation at 3500 × g for 10 min.Tissue samples (muscle and skin in natural proportions,minimum of 45 g) were collected in situ from euthanizedfish, placed in polyethylene bags, and immediately trans-ferred to the laboratory on dry ice. All samples were storedat −80 ◦C until analysis.

2.3. Analytical procedures

Erythromycin in plasma was analyzed by high per-formance liquid chromatography–electrospray mass spec-trometry (HPLC–ESI-MS) previously validated by CIDAS.A.L (Investigation and Applied Development Center S.A.L,Barcelona, Spain) according to the European Commis-sion Decision 2002/657/EEC (European Commission, 2002).The study of tissue depletion of erythromycin residues inrainbow trout was based on the European CommissionGuideline for studies intended to establish the withdrawalperiods of an antibiotic (European Commission, 2005).

Plasma and tissue samples were submitted to twoextractions with chloroform and subsequently analyzedby HPLC for the purification and quantification of ery-thromycin A residues. Samples were homogenized with4 ml of chloroform, vortexed horizontally for 10 min andcentrifuged at 4 ◦C for 5 min at 3000 × g. The supernatantwas removed and the procedure repeated. The combinedorganic phases were evaporated and the pellet resus-pended with 200 �l of the mobile phase. Finally, 25 �l of

the resulting solution was injected into the HPLC system,with a run time of 10 min.

The HPLC–ESI-MS analysis was performed using aWaters Micromass ZQ mass spectrometer and analyzed

Medicine 105 (2012) 160– 163 161

with the MassLynxTM software, version 4.0 (Micromass,USA). A mobile phase consisting of 4 mM aqueous ammo-nium acetate + 0.2% acetic acid/acetonitrile (60:40, v/v) wasused. The column temperature was maintained at 30 ◦C.The flow rate was 1.0 ml min−1. Nitrogen was used as des-olvation gas at a flow rate of 450 l h−1 and as cone gas ata flow rate of 50 l h−1. The desolvation temperature was250 ◦C. Capillary and cone voltages were 3000 and 20 V,respectively. The ionization source was used at 100 ◦C.

The method was previously validated through thedetermination of precision, accuracy, specificity, stability,calibration curve, and detection capability according to theCommission Decision 657/2002 (European Commission,2002). The precision of the method was evaluated by theinter- and intra-day (n = 6) assays at three different con-centrations of erythromycin A in the range from 10 ng ml−1

to 2.5 �g ml−1 (quality control plasma samples) and from100 �g kg−1 to 1000 �g kg−1 (quality control tissue sam-ples). The accuracy of the method obtained by qualitycontrol samples was studied by calculating the mean recov-ery of the target compound by adding standards at knownconcentrations to the samples. Specificity was evaluated bythe analysis of antibiotic-free samples (plasma or tissues)and applying the analytical procedure to determine theextent to which endogenous components may contributeto the interference at retention time of erythromycin Aand internal standard. A suitable volume of stock stan-dard solutions of erythromycin A was added to 0.1 ml ofantibiotic-free plasma to obtain a calibration curve rangingfrom 0 ng ml−1 to 5000 ng ml−1. Similarly, erythromycin Awas added to 250 mg of antibiotic-free tissue to obtain acalibration curve ranging from 0 �g kg−1 to 5000 �g kg−1.

2.4. Statistical analysis

The results are presented as medians and were com-pared using the Mann–Whitney U test or Kruskal–Wallistest as appropriate, because most data were not normallydistributed. Data were analyzed using SPSS for Windowsversion 17.0 (SPSS, Chicago, IL).

3. Results and discussion

The results indicated that the precision of the methodwas acceptable, with coefficients of variation less than15%. In addition, the chromatograms were found to befree of interfering peaks, which demonstrate the speci-ficity of the analytical procedure. The calibration curvesshowed a good linearity in the range of tested concen-trations (10–5000 ng ml−1 for the plasma analysis, and100–5000 �g kg−1 for the tissue analysis) with a correla-tion coefficient (r2) equal to 0.995.

The maximum levels of erythromycin in plasma(median 5.59 �g ml−1, 95% CI 2.18–12.09 �g ml−1)were obtained during the first day of treatment.The plasma erythromycin levels between days 2

and 8 oscillated between 3.33 and 1.28 �g ml−1

(median values). At the end of the treatment, we observeda fast decrease in plasma erythromycin concentrationson days 10, 12 and 14, reaching concentrations of 0.39,

162 D. Vendrell et al. / Preventive Veterinary Medicine 105 (2012) 160– 163

Table 1Plasma concentrations of erythromycin (�g ml−1) from rainbow trouttreated with medicated feed (52 mg erythromycin kg−1 trout bodyweight) for 8 consecutive days. Plasma samples were obtained daily 12 hafter treatment administration.

Time Erythromycinconcentration in plasma(�g ml−1)a

Day ◦C-days

0 0 ND1 18 5.59 (2.18–12.09)a2 36 3.33 (1.32–6.17)a,b3 54 2.21 (1.28–2.60)b4 72 2.28 (1.43–2.77)b5 90 1.28 (0.45–2.38)b6 108 2.10 (1.39–2.79)b7 126 1.46 (0.88–2.66)b8 144 1.61 (1.18–2.19)b

10b 180 0.39 (0.04–0.61)c12b 216 0.06 (0.00–0.33)c,d14b 252 0.00 (0.00–0.21)d

ND, not detected.D

0tdsfiiLdatMepMfi0g

rttt(LPtbls“(odtwsu

Table 2Erythromycin depletion at different times in tissue samples (muscle plusskin in natural proportion) from rainbow trout treated with medicatedfeed (52 mg erythromycin kg−1 trout body weight) for 8 consecutive days.Degree days are calculated by multiplying the mean daily water temper-ature by the total number of days measured.

Time (post-treatment) Erythromycinconcentration in muscleplus skin in naturalproportion (�g kg−1)b

Day ◦C-days

2 36 1173.83 (397.14–1274.45)4 72 248.60 (125.41–402.30)6 108 220.90 (145.50–296.30)8 144 <LOQa

10 180 <LOQ15 270 <LOQ20 360 <LOQ28 504 <LOQ

ifferent letters denote significant differences (P < 0.05).a Values are expressed as medians with confidence intervals (95% CI).b Days 10, 12 and 14 were 2nd, 4th and 6th day post-treatment.

.06 �g ml−1 (median values) and not detectable, respec-ively (Table 1). Statistical analysis revealed that theecreased plasma erythromycin concentrations wereignificant (P < 0.05) when compared with those of treatedsh. These data were then compared to the minimum

nhibitory concentration (MIC) of erythromycin against. garvieae in order to determine whether the proposedosage is suitable to control the disease. Although therere differences between plasma levels of sampled fish athe same time point, all individual scores were above the

IC90 for L. garvieae (MIC90 = 0.125 �g ml−1) (Vendrellt al., 2008). In fact, the minimum score obtained inlasma for an individual sample was three times theIC90. The minimum erythromycin concentration for 10

sh was obtained on day 5 (median 1.28 �g ml−1, 95% CI.45–2.38 �g ml−1); this value is ten times the MIC90 for L.arvieae.

With regard to the tissue depletion of the antibiotic,esults showed that concentrations of erythromycin inissue samples were quite low (Table 2). At the thirdime point (6 days post-treatment) the concentrations ofhe antibiotic were lower than the limit of quantificationLOQ = 100 �g kg−1) in many of the individual samples.inear regression (Software Withdrawal-Time Calculationrogram), as recommended by the European Agency forhe Evaluation of Medicinal Products (EMEA, 1996), cannote applied to estimate the withdrawal time because of the

ow concentrations of erythromycin observed in the tis-ue samples. However, according to the note for guidance:Approach towards harmonization of withdrawal periods”EMEA, 1996), when results do not allow the applicationf linear regression, an alternative proposal based on theetermination of the first time point where the concen-

ration of erythromycin residues for all sampled animalsas below the MRL for erythromycin (MRL = 200 �g kg−1)

hould be considered as established by the Council Reg-lation no. 37/2010 EC (European Commission, 2010).

a LOQ (limit of quantification): 100 �g kg−1.b Values are expressed as medians with confidence intervals (95% CI).

According to this proposal, the addition of a safetyfactor of 10–30% should also be considered. We thus choseto use the maximum safety factor recommended. The firsttime point where all scores of erythromycin A were below200 �g kg−1 was 8 days after the end of treatment. Afteradding 30% of 8 days, we obtained an estimated withdrawaltime of 10.4 days. The mean daily water temperature dur-ing all the experimental period, an important factor on fishmetabolism and, consequently, on the pharmacokineticsof the antibiotic, was 18 ± 1 ◦C. Therefore, the withdrawaltime was of 187 ◦C-day (◦C-day: mean water tempera-ture × days).

4. Conclusions

This study suggests that erythromycin administration atthe recommended dosage of 52 mg kg−1 body weight day−1

during 8 consecutive days is appropriate to obtain thera-peutic levels in plasma against L. garvieae. The withdrawaltime determined for the antibiotic (187 ◦C-day) is belowthe general withdrawal time of 500 ◦C-day recommendedby the European Commission for veterinary medicinalproducts authorized in the Member State under the Direc-tive 2004/28/EC or Regulation (EC) no. 726/2004. Therefore,these data may be useful to establish an adequate therapeu-tic regimen in fish.

Acknowledgments

This work was supported by Laboratorios HIPRA, Spain.We thank J.A. Nienow for critical reading of the manuscript.

References

EMEA, 1996. Note for guidance on approach towards harmonisationof withdrawal periods. http://www.ema.europa.eu/docs/en GB/document library/Scientific guideline/2009/10/WC500004428.pdf

(accessed 23.08.10).

Esposito, A., Fabrizi, L., Lucchetti, D., Marvasi, L., Coni, E., Guan-dalini, E., 2007. Orally administered erythromycin in rainbow trout(Oncorhynchus mykiss): residues in edible tissues and withdrawaltime. Antimicrob. Agents Chemother. 51, 1043–1047.

terinary

D. Vendrell et al. / Preventive Ve

European Commission, 2010. Commission Regulation no. 37/2010 of 22December 2009 on pharmacologically active substances and theirclassification regarding maximum residue limits in foodstuffs of ani-mal origin. Off. J. Eur. Union L15, 1–72.

European Commission, 2005. Establishment of maximum residue limits(MRLs) for residues of veterinary medicinal products in food-stuffs of animal origin. http://ec.europa.eu/health/files/eudralex/vol-8/pdf/vol8 10-2005 en.pdf (accessed 23.08.10).

European Commission, 2002. 2002/657/EC: Commission decision of 12August 2002 implementing Council Directive 96/23/EC concerning the

performance of analytical methods and the interpretation of results.Off. J. Eur. Communities L221, 8–36.

Fairgrieve, W.T., Masada, C.L., Peterson, M.E., McAuley, W.C., McDow-ell, G.C., Strom, M.S., 2006. Concentrations of erythromycin andazithromycin in mature Chinook salmon Oncorhynchus tshawytscha

Medicine 105 (2012) 160– 163 163

after intraperitoneal injection, and in their progeny. Dis. Aquat. Organ.68, 227–234.

Munday, B.L., 1994. Fish. In: Coopert, S.B. (Ed.), Antimicrobial PrescribingGuidelines for Veterinarians. University of Sydney in association withthe National Health and Medical Research Council, Sydney, Australia,pp. 305–325.

Vendrell, D., Serarols, L., Balcázar, J.L., de Blas, I., Ruiz-Zarzuela, I., Múzquiz,J.L., 2008. Minimum inhibitory concentrations of erythromycin inLactococcus garvieae strains isolated from cultured rainbow trout(Oncorhynchus mykiss) in Spain. Bull. Eur. Assoc. Fish Pathol. 28,

125–128.

Vendrell, D., Balcázar, J.L., Ruiz-Zarzuela, I., de Blas, I., Gironés, O., Múzquiz,J.L., 2007. Safety and efficacy of an inactivated vaccine against Lac-tococcus garvieae in rainbow trout (Oncorhynchus mykiss). Prev. Vet.Med. 80, 222–229.