Prioritisation of Veterinary Medicines in the UK Environment

Home , Ectoparasiticide

Toxicology Letters 142 (2003) 207Á/218 www.elsevier.com/locate/toxlet Short communication Prioritisation of veterinary medicines in the UK environment

Alistair B.A. Boxall a,*, Lindsay A. Fogg a, Paul Kay a, Paul A. Blackwel1 a, Emma J. Pemberton b, Andy Croxford b

a Cranﬁeld Centre for EcoChemistry, Cranﬁeld University, Shardlow Hall, Shardlow, Derby, DE72 2GN, UK b Environment Agency, National Centre for Ecotoxicology and Hazardous Substances, Evenlode House, Howbery Park, Wallingford, Oxon, OX10 8BD, UK

Received 22 July 2002; received in revised form 7 October 2002; accepted 8 October 2002

Abstract

A wide range of veterinary medicines is used to treat animals in the UK. Whilst the environmental impact of selected substances (particulary the sheep dip chemicals, anthelmintics and fish farm chemicals) has been well studied, limited information is available in the public domain on the other groups of substances (e.g. antifungals, coccidiostats, antiprotozoals, hormones and growth promoters). There is therefore a need to identify other substances that may impact the environment in order to design national monitoring programmes, target experimental work and develop pollution prevention methodologies. In this study, a simple two-stage prioritisation scheme was developed and applied to veterinary medicines in use in the UK. In the first stage, those substances that have high potential to enter the environment in significant amounts were identified on the basis of amounts used in the UK, treatment type and metabolism. In stage 2, the hazard of the identified substances to terrestrial and aquatic organisms was assessed. Using the approach, a total of 56 substances or groups were assigned to a ‘high priority’ category. For eleven of these substances, sufficient data were available to characterise their risk, these were: oxytetracycline, chlortetracycline, tetracycline, sulphadiazine, amoxicillin, diazinon, tylosin, dihydrostreptomycin, apramycin, cypermethrin and sarafloxicin. For the remaining 45 substances, full datasets were not available and it is recommended that in the first instance, attempts are made to fill these data gaps. # 2003 Elsevier Science Ireland Ltd. All rights reserved.

Keywords: Veterinary medicine; Environmental risk; Prioritisation; Monitoring; Pollution

1. Introduction tives (growth promoters) are also incorporated Veterinary medicines are widely used in the UK into the feed of animals reared for food in order to and across Europe to treat disease and protect the improve their growth rates. Under Directive 81/ health of animals. Dietary enhancing feed addi- 852/EEC as amended by 92/18/EEC veterinary medicinal products must be assessed for their quality, efficacy and safety (to both humans and * Corresponding author. the environment). Only products approved for use E-mail address: [email protected] (A.B.A. Boxall). by the regulatory authority may be used.

Release of veterinary medicines to the environ- EA to: (1) guide policy direction; (2) ensure that ment occurs both directly, for example the use of their monitoring programme is effectively tar- medicines in fish farms, and indirectly, via the geted; and (3) identify the need for pollution application of animal manure (containing excreted prevention measures. products) to land. A number of groups of veterinary medicines, primarily sheep dip chemicals (Environment Agency, 1998, 2000, 2001; SEPA, 2. Prioritisation approach 2000), fish farm medicines (Davies et al., 1998; Jacobsen and Berglind, 1988) and anthelmintics 2.1. Collation of data (McKellar, 1997; Strong, 1993; McCracken, 1993; Madsen et al., 1990; Ridsill-Smith, 1988; Wall and Data on amounts and/or sales of veterinary Strong, 1987)have been extensively studied and a medicines in the UK were obtained from a number large body of data is available for these substances. of sources, including: (1) data obtained from Information is also available on selected antibio- Intercontinental Medical Statistics (IMS) Health tics (e.g. Halling-Sørensen, 1999; Holten Lu¨tzhøft and summarised in Boxall et al. (2002a); (2) the et al., 1999). However, there are scant data Veterinary Medicines Directorate (VMD) data on available in the public domain on the environ- the sales of antimicrobial substances and sheep dip mental fate, behaviour and effects of other generic chemicals in the UK (VMD, 2001); and (3) data in groups of veterinary medicines and so their the published literature on the use of sheep dip potential environmental impacts are less well chemicals (Liddel, 2000). Information on the understood (Jørgensen and Halling-Sørensen, metabolism, usage pattern and ecotoxicity of 2000). veterinary medicines in use in the UK were also The large number and wide variety of veterinary obtained, detailed information is reported in medicines available means that it is difficult to Boxall et al. (2002a). identify those substances that should be included in National monitoring programmes and those 2.2. Outline of prioritisation approach substances that should be investigated in further detail in terms of environmental fate and effects. The prioritisation exercise considered data on There is therefore an urgent need to identify those usage, exposure routes and environmental effects substances that are likely to have the greatest of all generic groups of veterinary medicines. As potential to impact the environment. If this could the focus of the study was on potential environ- be achieved, then future monitoring programmes mental impacts, the issue of microbial resistance, and experimental studies could be targeted at which may threaten the future effectiveness of substances of concern. antibiotic treatments for livestock and humans, The impact of a veterinary medicine on the was not considered. An overview of the prioritisa- environment will be determined by a range of tion process is illustrated in Fig. 1. The prioritisa- factors including, the quantity used, the degree of tion process was performed in two stages, further metabolism in the animal and degradation during details are provided below. storage of manure prior to land spreading and the toxicity of the substance to terrestrial and aquatic 2.3. Stage 1. Potential to reach the environment in organisms. This paper describes the application of signiﬁcant amounts a straightforward prioritisation scheme that in- corporates these factors, for identifying veterinary Using data on usage, pathways of entry to the medicines that have the potential to impact the environment and metabolism, those veterinary UK aquatic and terrestrial systems. The scheme medicines considered to reach the environment in has been applied, by the Environment Agency of potentially significant amounts were identified. England and Wales (EA), to veterinary medicines Groups of substances were initially ranked as in use in the UK. The results will be used by the high (]/10 tonnes per annum (tpa)), medium A.B.A. Boxall et al. / Toxicology Letters 142 (2003) 207Á/218 209

Fig. 1. Schematic presentation of the prioritisation process used. 210 A.B.A. Boxall et al. / Toxicology Letters 142 (2003) 207Á/218

(]/1Á/ B/10 tpa), low (B/1 tpa) or unknown usage, groups, the potential to reach the environment in using the compiled data. The potential for the significant amounts is considered high when used substance to enter the environment was then in aquaculture but low when used to treat indivi- assessed using information on: (1) the target duals. Likewise, compounds are classified as hav- treatment group; (2) route of administration; (3) ing a higher potential to enter the environment metabolism in the animal; and (4) the potential for when used as topical herd treatments than when the substance to be degraded in slurry or manure used topically to treat companion animals or during storage. Substances were classified as hav- individuals. ing high, medium, low or unknown potential to enter the environment using the criteria detailed in 2.4. Stage 2. Hazard assessment Table 1. Using the classifications determined for usage For those compounds that were identified as and potential to enter the environment, those having the potential to enter the environment in substances considered to have the greatest poten- significant quantities, a simple assessment of tial to enter the environment and therefore requir- hazard was conducted using the toxicity data ing hazard assessment, were identified using the provided in Boxall et al. (2002a). This enabled matrix detailed in Table 2. Compounds identified identification of those compounds having a high as both high usage and having a high potential to potential to enter the environment and which were enter the environment were considered to poten- the most toxic (and thus represented potentially tially represent the highest risk to the environment the highest risk to the environment). These com- and hence were deemed to be the highest priority pounds were considered to be the highest priority for further assessment. for further consideration of their impact on the For those compounds regarded as having low environment and the possible need for control potential to enter the environment it was consid- measures such as pollution reduction programmes. ered unnecessary to assess their intrinsic hazard in Substances were classified as having very high, the prioritisation exercise, as relative to the other high, medium or low aquatic and/or terrestrial veterinary medicines, they are likely to represent a ecotoxicity using the criteria detailed in Table 3. low risk to the environment. This group included The hazard classification ‘unknown’ was assigned those compounds administered either orally or by to those compounds where no data for aquatic injection (non-topical applications) as herd treat- toxicity or terrestrial toxicity was available. ments that are significantly metabolised, as well as As an indication of the relative completeness of compounds used to treat companion or individual the available data on which the hazard classifica- food production animals by non-topical routes. In tion was determined, a score was assigned (these addition, compounds with a medium potential to are given in the footnote of Table 5). For aquatic enter the environment, for example those used as hazard classifications, the score took into account herd treatments that are moderately metabolised the number of trophic levels tested as well as the as well as those used to treat companion or type of tests conducted. Chronic tests for three individual animals by application to the skin different trophic levels were regarded as being (topical applications) were excluded from hazard more comprehensive than a mixture of chronic assessment when usage was less than one tonne per tests for one or two trophic levels and several acute annum. toxicity tests. A simpler system was adopted for For compounds that are used on more than one the terrestrial data than for aquatic toxicity data target treatment group, the potential to reach the because there were comparably fewer toxicity data environment was assessed separately for each available for terrestrial species. target group, i.e. companion/individuals, herds Considering both the potential to reach the and aquaculture, as this may affect the potential environment (stage one) and hazard classification for environmental impact. For example, for com- (stage two) substances were then assigned to one pounds that are used to treat all three target of five groups using the matrix detailed in Table 4. Table 1 207 (2003) 142 Letters Toxicology / al. et Boxall A.B.A. Criteria used to assess the potential for the environment to be exposed to an individual veterinary medicine

Classification Target Route of Metabolism Rationale group administration

High Aquaculture Topical/other na Substances typically applied directly into the aquatic environment. Herd Topical na As the substances are applied topically, there is the potential for wash-off from the animal. Topical treatments used in herds are likely to enter the environment in higher amounts than topical treatments used to treat individual or companion animals because of the quantities used. Herd Other L Potential impact from substances used as herd treatments that are not significantly metabolised. Medium Herd Other M Potential impact from substances used as herd treatments that are moderately metabolised. Companion/ Topical na Potential for direct entry to the environment in excreta. However since only individuals are treated the individual environmental impact is considered to be lower than for herd treatments. Topical treatments have a higher potential to reach the environment than ‘other’ routes of administration. Low Herd Other H Low potential for substances used as herd treatments to enter the environment because of significant metabolism. Companion/ Other na Negligible environmental impact on the basis that it is individuals that are treated rather than herds, therefore individual metabolism is not considered. Unknown Herd Other U Unknown potential to enter environment because of insufficient data on metabolism. Á / 218 Metabolism: H/ /80%; M/20Á/80%; L/ B/20%; U/unknown; na/not applicable; Other/orally or by injection. Individual/individual food production animals. 211 212 A.B.A. Boxall et al. / Toxicology Letters 142 (2003) 207Á/218

Table 2 Matrix used to identify substances requiring hazard assessment

Usage Potential to enter Hazard assessment required? environment combinations

HH Y M All other

HM Y b /L

HL N b

HU Y /M b MH Y MM Y ML N MU Y M H/U/H

LH Y b

LM N /L b LL N /M LU Y b UH Y UM Y UL N UU Y

H/high; M/medium; L/low; U/unknown. L M/L H/U/H b

Table 3 /L Classiﬁcation criteria for ecotoxicity b /M b Hazard Aquatic toxicitya Terrestrial toxicityb classification (mg l1) (mg kg1)

VH 5/0.1 5/10 H /0.15/1 /105/100 M H/U/H M /15/100 /1005/1000 b

L /100 /1000 /L b /M VH/very high; H/high; M/medium; L/low. b

a unknown. Based on harmonised system for the classification of / chemicals which are hazardous for the aquatic environment; OECD (1998). M/U/H b b low; U /L Based on a proposed EU hazard assessment scheme for the / b terrestrial environment. /M b Compounds assigned to group one were consid- H/U/H

ered to have the greatest potential for environ- medium; L / b

mental impact and thus are the highest priority for /L further work. These were compounds that had a b /M b

combination of high or medium usage, together high; M / with high or medium potential to enter the environment and very high or high toxicity to Priority classification 12 34 H/M/U H/U H/U M M H/U H/U H/U M M H/U either aquatic or terrestrial organisms. Com- pounds that were considered to have low potential ery high; H v to enter the environment in significant amounts / Usage data incomplete. and thus did not require a hazard assessment were ironment b VH v Potential to enter en Usage H/U/H Table 4 Matrix used to determine the priority classiﬁcation of a substance assigned to the lowest score of 5. Where there was Hazard VH/U H VH/U VH/U H VH/U H M H M M A.B.A. Boxall et al. / Toxicology Letters 142 (2003) 207Á/218 213 uncertainty in any one of the three criteria used, There was only sufficient data available to fully such as unknown data (U) or in the case of usage, characterise the potential risk for eleven of these incomplete data, the worst case classification was compounds (Table 6). For two of these substances assumed. (oxytetracycline and amoxicillin), the classification was obtained for both herd and aquaculture treatment scenarios and for two substances (su- 3. Results phadiazine and sarafloxicin), the classification was obtained for the aquaculture scenario. The re- After Stage 1 a number of therapeutic groups maining substances (chlortetracycline, tetracycline, were identified that were considered to have diazinon, tylosin, dihydrostreptomycin, apramycin sufficiently low potential to enter the environment and cypermethrin) were assigned to the high that they did not require a hazard assessment. priority class as a result of their use as herd These included general anaesthetics for companion treatments. For the remaining 45 compounds animals and therapeutic groups where usage was some of the data required for the prioritisation less than 1 tonne per annum (tpa) therapeutic exercise were either unavailable or incomplete and groups (where individual compounds were not so the prioritisation exercise has incorporated one identified). Fifteen individual substances from or more worst-case assumptions. Compounds other groups were excluded from further assess- identified as potentially high risk (group one), ment, these included some compounds that were but requiring further data are also shown in Table considered to be high usage but which had a high 6. potential for metabolism (including sulphadimi- Six compounds were assigned to group two dine, dimetridazole, narasin and avilamycin). (Table 6). These compounds are considered to The general anaesthetics are typically gaseous potentially represent a risk to the environment, but and are hence unlikely to reach water or land in are of less concern than the group one compounds significant quantities. Furthermore, the release of discussed above. None of these compounds had gaseous compounds to the atmosphere will be complete data sets for the purposes of the prior- subject to significant dissipation in air and as a itisation exercise. result aerial exposure is likely to be minimal. Therapeutic groups where usage was less than 1 tpa included some antifungals, neurological pre- 4. Discussion parations and anti-inflammatory preparations. Several other therapeutic groups were also con- A pragmatic and scientific approach has been sidered as low priority despite usage being un- developed in order to enable an initial identifica- known because they used to treat individual tion and prioritisation of those veterinary medi- animals (companion or food production). These cines of environmental concern to be made, using included the anti-inflammatory steroids, diuretics, available data. The exercise has identified those cardiovascular and respiratory treatments and compounds considered to have the greatest poten- locomotor treatments. tial to cause environmental impacts as a conse- Compounds identified as having a high poten- quence of their use. However, it is important to tial to enter the environment and of high usage recognise that many compounds identified as high included a number of antimicrobial compounds priority in this exercise may not actually cause (the tetracyclines, sulphadiazine, trimethoprim, adverse impacts on the environment. The prior- amoxicillin, tylosin, dihydrostreptomycin, neomy- itisation exercise is simply a way of assessing the cin and apramycin) and diazinon, an ectoparasiti- relative potential for veterinary medicines to cause cide commonly used in sheep dip preparations harm, thus enabling those compounds likely to be (Table 5). Hazard assessment of these substances of greatest concern to be identified and monitored. resulted in a total of 56 compounds being assigned For those compounds where sufficient data was to the ‘high priority’ category (group one). available, the list provides a system of relative 214 Table 5 Prioritisation assessment for veterinary medicinal products that have the potential to enter the environment

Therapeutic group Chemical group Major usage products Potential to reach Relevant target Usage Hazard assessment Priority (where data available) environment group(s) class classification Aquaticc Terrestriald

Antimicrobials Tetracyclines Oxytetracycline H H, A H H3 L3 1 Chlortetracycline H H VH4 VH3 1 Tetracycline H H VH4 U1 Antimicrobials Potentiated Sulphadiazine H A H H4 H3 1 sulphonamides ...Bxl ta./Txclg etr 4 20)207 (2003) 142 Letters Toxicology / al. et Boxall A.B.A. Trimethoprim H A M4 U1 Baquiloprim U H U U 1 a b Endoparasiticides Á/ Amprolium MHHUVH3 1 -coccidiostats Clopidola UH UU1 Lasalocid sodiuma UH UU1 a Maduramicin MH UVH2 1 Nicarbazina UH UU1 Robenidine hydrochloridea UH UU1

Antimicrobials b-lactams Amoxicillin H H, A H VH4 U1 Procaine penicillin U H U VH3 1 Procaine benzylpenicillin U H VH4 U1 Clavulanic acid U H U U 1

Ectoparasiticides Organophosphates Diazinon H H H VH4 VH3 1 -sheep dips

Antimicrobials Macrolides Tylosin H H H VH4 L3 1 b Growth promoters Á/ Monensin U H H UVH2 1 a Salinomycin sodium UH UVH2 1 Flavophospolipola UH UU1

Antimicrobials Aminoglycosides Dihydrostreptomycin H H H VH4 U1

Neomycin H C, H L4 U1 Á / 218 Apramycin H H U VH1 1 Flavomycina UH UU1 Endoparasiticides Pyrimidines Morantel M H Mb UU 1 -wormers

Ectoparasiticides Pyrethroids Cypermethrin H H M VH4 U1 -sheep dips Flumethrin H H U U 1 Endoparasiticides Azoles Triclabendazole M H Mb UU 1 -wormers Fenbendazole U H U U 1 Levamisole U H U U 1 b Endoparasiticides Macrolide endectins Ivermectin M H M VH3 VH2 1 -wormers b Antimicrobial Á/ Cephalexin U H M UU 1

-other antibiotics Florfenicol H A U VH3 1 Table 5 (Continued )

Tilmicosin U H U U 1 a Oxolinic acid HA VH4 U1 b Neurological preparations Á/ Procaine hydrochloride U H M M4 U2 -local anaesthetics Lido/lignocaine hydrochloride U H U U 1 b Antimicrobials Pleuromutilins Tiamulin U H M VH3 M2 1 Antimicrobials Lincosamides Lincomycin U H M M4 VH1 1

Clyndamycin U H U U 1 207 (2003) 142 Letters Toxicology / al. et Boxall A.B.A. Antimicrobials Azoles Miconazole M C M U U 2 -antifungals Endoparasiticides Others Nitroxynil U H Mb UU 1 -wormers Antimicrobials Fluoroquinolones Enrofloxacin H H M U U 1

Sarafloxacin A VH4 VH1 1 Sex hormones Á/ Altrenogest U H L U U 2 Progesterone U H U U 2 Medroxyprogesterone U H U U 2 b Enteric preparations Á/ Dimethicone U H L UU 1 Poloxalene U H U U 1 b Endoparasiticides Á/ Toltrazuril U H L UU 1 -antiprotozoals Decoquinate U H U U 1 Diclazuril U H U U 1 Endectocides Macrocyclic lactone injections Moxidectin U H L U U 2 b Ectoparasiticides Á/ Phosmet H H U/L UU 1 -others Piperonyl butoxide M C U U 1

Ectoparasiticides Amidines Amitraz H H U M2 U1 -sheep dips Á / Ectoparasiticides Á/ Deltamethrin H H U VH4 H3 1 218 -spray and pour-ons for sheep Cypromazine H H VH4 U1 Ectoparasiticides Á/ Emamectin benzoate H A U VH4 na 1 -aquaculture treatments

Antiseptics Á/ ?HC/IUUU1

Immunological products Á/ ?UC,HUUU1

a Specific usage data unavailable, however compound considered to be potentially major usage. b Usage data incomplete/ c Aquatic scores: subscript1, 3 trophic levels, chronic test; subscript2, 3 trophic levels, acute or chronic test; subscript3, 3 trophic levels, acute test; subscript4, less than 3 trophic levels, acute or chronic test or both. d Terrestrial scores: subscript1, 3 trophic levels; microbes, invertebrate and plants; subscript2, any 2 of 3 trophic levels; subscript3,Á/ any 1 of 3 trophic levels. 215 216 A.B.A. Boxall et al. / Toxicology Letters 142 (2003) 207Á/218 ranking on the basis of potential environmental Table 6 impact. Eleven substances were assigned to group Substances assigned to Groups 1 and 2 during the prioritisation one, on the basis of a ‘complete’ data set and thus exercise considered to be the highest priority. These sub- Group 1 substances stances include a number of antimicrobials widely 1. oxytetracycline 5. amoxicillin (H,A) 9. cypermethrin (H) used as herd treatments and/or in aquaculture (H,A) (oxytetracycline, chlortetracycline, tetracycline, 2. chlortetracy- 6. diazinon (H) 10. sarafloxicin (A) cline (H) sulphadiazine, amoxicillin, tylosin, dihyrostrepto- 3. tetracycline 7. tylosin (H) mycin and apramycin). A further antimicrobial (H) compound, sarafloxacin, used exclusively in aqua- 4. sulphadiazine 8. dihydrostreptomy- culture treatments, was also identified as a high (A) cin (H) priority as were diazinon and cypermethrin, two Group 1 possible substances compounds used extensively in sheep dips. 1. trimethoprim 17. morantel 33. dimethicone Both cypermethrin and diazinon are known to 2. baquiloprim 18. flumethrin 34. poloxalene 3. amprolium 19. triclabendazole 35. toltrazuril cause environmental pollution and a significant 4. clopidol 20. fenbendazole 36. decoquinate body of data on their environmental fate, beha- 5. lasalocid so- 21. levamisole 37. diclazuril viour and ecotoxicity is available. Pollution in- dium cidents caused by poor sheep dipping practises can 6. maduramicin 22. ivermectin 38. phosmet result in ecological damage over several kilometres 7. nicarbazin 23. cephalexin 39. piperonyl butoxide of watercourse (e.g. SEPA, 2000). Sheep dip 8. robenidine hy- 24. florfenicol 40. amitraz chemicals are routinely monitored and in the UK drochloride each year there are a relatively high number of 9. procaine peni- 25. tilmicosin 41. deltamethrin sites failing the Environmental Quality Standards cillin (EQS) (which are derived, using appropriate un- 10. procaine ben- 26. oxolinic acid 42. cypromazine zylpenicillin certainty factors, from available ecotoxicity data) 11. clavulanic 27. lido/ligocaine 43. emamectin for both cypermethrin and diazinon (Environment acid HCL benzoate Agency, 2000, 2001). 12. monensin 28. tiamulin 44. antiseptics However, with the exception of a few studies 13. salinomycin 29. lincomycin 45. immunological (e.g. Kolpin et al., 2002; Hamscher et al., 2001; sodium products 14. flavophospo- 30. clindamycin Boxall et al., 2002b) the chemicals (other than lipol cypermethrin and diazinon) identified as a high 15. neomycin 31. nitroxynil priority have not been looked for in the environ- 16. flavomycin 32. enrofloxacin ment and only a few published studies have Group 2 substances investigated environmental effects (e.g. Halling- 1. procaine HCL 3. altrenogest 5. medroyprogester- Sørensen, 1999; Holten Lu¨tzhøft et al., 1999; one Wollenberger et al., 2000). Further assessment 2. miconazole 4. progesterone 6. moxidectin and limited targeted monitoring is therefore re- The treatment scenario giving rise to a potential risk to the commended to ascertain whether these chemicals environment is indicated in parentheses (H/herd treatment; are present in the environment at ecologically A/aquaculture treatment). significant levels. Ideally, this would involvean integrated chemical and biological monitoring programme. a high priority, including many other antimicro- The prioritisation exercise highlighted the fact bial, coccidiostat, endo- and ectoparasiticide, anti- that there are many veterinary medicines for which fungal, antiprotozoal and growth promoting little or no data are available in the public domain. substances. However, for many of these com- Classification of many of the compounds was pounds either accurate usage information was based on limited data and worst case assumptions. unavailable or their potential to enter the environ- Forty-five substances were provisionally ranked as ment or intrinsic hazard was unknown. It is A.B.A. Boxall et al. / Toxicology Letters 142 (2003) 207Á/218 217 considered a priority for any future work that data 5. Conclusions should be obtained for these compounds in order to refine and extend the current work. This is There are a large number of veterinary medi- required in order to ascertain whether such cines in use in the UK. Whilst the concentrations chemicals are correctly classified in terms of their and/or behaviour and effects of selected groups of potential risk to the environment in the current veterinary medicines has been well characterised, exercise. Those that have been correctly classified limited information is available on the potential can then be added to the list of 11 substances impacts of the other substances. A scientifically described above for further consideration of their sound and pragmatic approach has therefore been environmental impact. developed for identifying substances that may pose It should be recognised that the work has a risk to terrestrial and aquatic systems in the UK. focused exclusively on the parent compound. The approach has been applied using information However, following injection or oral administra- on tonnage sold, typical usage regimes, metabo- tion to an animal, compounds may be metabolised lism and toxicity to aquatic and terrestrial organ- and subsequently excreted, in part or completely, isms. Eleven substances, including antibiotics and as transformation products. In addition, if ex- ectoparasiticides, have been identified as high creted as the unaltered parent compound they may priority and a further 45 substances have been degrade on reaching the environment. The poten- identified as potentially high priority but requiring tial environmental impact of any metabolites or further data. It is recommended that in the future, the data gaps are addressed and that the high degradation products should be assessed, espe- priority substances are further assessed. Targeted cially for those compounds considered to be low monitoring and fate and effects studies should priority on the basis of this prioritisation exercise then be performed to determine the impacts, if because they are extensively metabolised following any, that these substances may be having on the administration. Data on metabolism and environ- environment. mental degradation were very limited, and conse- quently detailed consideration in the prioritisation exercise was not possible. Several veterinary medicines, for which there Acknowledgements were no usage data, were included in the prioritisation exercise as they may be distributed via The authors would like to thank the UK routes other than those covered by the information Environment Agency for funding this work. Part available to this study and they are therefore of the review phase of the study was performed potentially major usage compounds. Likewise, during the EU Framework V Project ERAVMIS two therapeutic groups (antiseptics and immuno- (project number EVVK1-CT-1999-00003) and the logical compounds) for which individual com- authors would like to thank the European Com- pounds have not been identified are also mission for their ﬁnancial support. included, on the basis that there could also be major usage compounds. Whilst the prioritisation exercise has focused on the UK situation, other studies (e.g. Jørgensen and References Halling-Sørensen, 2000; Pelicaan et al., 2001) indicate that many of the substances identified Boxall, A.B.A., Fogg, L., Blackwell, P.A., Kay, P., Pemberton, during this study are also high usage in other E.J., 2002. Review of veterinary medicines in the environ- countries. It is therefore likely that the results of ment. R&D Technical Report P6-012/8TR, UK Environ- this study could be usefully used to design ment Agency, Briston, UK, p. 257. Boxall, A.B.A., Blackwell, P., Cavallo, R., Kay, P., Tolls, J., monitoring programmes and set priorities in 2002b. The sorption and transport of a sulphonamide countries other than the UK. antibiotic in soil systems. Toxicol. Lett. 131, 19Á/28. 218 A.B.A. Boxall et al. / Toxicology Letters 142 (2003) 207Á/218

Davies, I.M., Gillibrand, P.A., McHenery, J.G., Rae, G.H., McCracken, D.I., 1993. The potential for avermectins to affect

1998. Environmental risk of ivermectin to sediment-dwell- wildlife. Vet. Parasitol. 48, 273Á/280. ing organisms. Aquaculture 163, 29Á/46. McKellar, Q.A., 1997. Ecotoxicology and residues of anthel-

Halling-Sørensen, B., 1999. Algal toxicity of antibacterial mintic compounds. Vet. Parasitol. 72, 413Á/435. agents used in intensive farming. Chemosphere 40, 731Á/739. Madsen, M., Overgaard-Nielsen, B., Holter, P., Pedersen, O.C., Hamscher, G., Sczesny, S., Hoper, H., Nau, H., 2001. Brocchener-Jespersen, J., Vagn Jensen, K.-M., Nansen, P., Determination of persistent tetracycline residues in soil Grovold, J., 1990. Treating cattle with ivermectins: effects fertilized with liquid manure by high-performance liquid on fauna and decomposition of dung pats. J. Appl. Ecol. 27, chromatography with electrospray ionization tandem mass 1Á/15. spectrometry. Anal. Chem. 74 (7), 1509Á/1518. OECD, 1998. Harmonised integrated hazard classification for Holten Lu¨tzho¨ft, H.C., Halling-Sørensen, B., Jorgensen, S.E., human health and environmental effects of chemicals. As 1999. Algal toxicity of antibacterial agents applied in endorsed by the 28th joint meeting of the Chemicals Danish fish farming. Arch. Environ. Contam. Toxicol. 36, Committee and the Working Party on Chemicals in 1 /6. Á November 1998. OECD, Paris. Environment Agency, 1998. Pesticides 1998: A summary of Pelicaan, C.H.P., van Turnhout, Pijpers, A., 2001. Verbruiks- monitoring of the aquatic environment in England and cijfers van antibacteriele middelen bij landbouwhuis diren. Wales. National Centre for Ecotoxicology and Hazardous Poster obtained from the University of Utrecht, The Substances, Environment Agency, Wallingford, UK. Netherlands. Environment Agency, 2000. Welsh sheep dip monitoring Ridsill-Smith, T.J., 1988. Effects of avermectin residues in cattle programme. Environment Agency, Cardiff, Wales. dung on dung beetle (Coleoptera: Scarabaeidae) reproduc- Environment Agency, 2001. Pesticides 1999/2000: A summary of monitoring of the aquatic environment in England and tion and survival. Vet. Parasitol. 48, 127Á/137. Wales. National Centre for Ecotoxicology and Hazardous SEPA, 2000. Long term biological monitoring trends in the Tay Substances, Environment Agency, Wallingford, UK. system 1988Á/1999. Scottish Environmental Protection Jacobsen, P., Berglind, L., 1988. Persistence of oxytetracycline Agency, Eastern Region, Scotland. Strong, L., 1993. Overview: the impact of avermectins on in sediment from fish farms. Aquaculture 70, 365Á/370. Jørgensen, S.E., Halling-Sørensen, B., 2000. Drugs in the pastureland ecology. Vet. Parasitol. 48, 3Á/17. VMD, 2001. Sales of antimicrobial products used as veterinary environment. Chemosphere 40, 691Á/699. Kolpin, D.W., Furlong, E.T., Meyer, M.T., Thurman, E.M., medicines and growth promoters in the UK in 1999. Zaugg, S.D., Barber, L.D., Buxton, H.T., 2002. Pharma- Veterinary Medicines Directorate, UK. ceuticals, hormones and other organic wastewater contami- Wall, R., Strong, L., 1987. Environmental consequences of

nants in US streams, 1999Á/2000: a national reconnaissance. treating cattle with the antiparasitic drug ivermectin. Nature Environ. Sci. Technol. 36, 1202Á/1211. 327, 418Á/421. Liddel, J.S., 2000. Sheep ectoparasiticide use in the UK: 1993, Wollenberger, L., Halling-Sørensen, B., Kusk, K.O., 2000. 1997 and 1999. Paper presented to the 5th International Acute and chronic toxicity of veterinary antibiotics to

Sheep Veterinary Congress, Stellenbosch, South Africa. Daphnia magna. Chemosphere 40 (7), 723Á/730.