leaching behaviour of pharmaceuticals in soil-testing-systems: a part of an environmental risk...

9
Science of the Total Environment 328 (2004) 265–273 0048-9697/04/$ - see front matter 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2004.02.004 Leaching behaviour of pharmaceuticals in soil-testing-systems: a part of an environmental risk assessment for groundwater protection J. Oppel , G. Broll , D. Loffler , M. Meller , J. Rombke *, Th. Ternes a b c d d, c ¨ ¨ ENVIRON Germany GmbH, Muhlwiese 9, D-65779 Kelkheim, Germany a ¨ Hochschule Vechta, Abt. F. Geo- und Agrarokologie (ISPA), Postfach 1553, D-49364 Vechta, Germany b ¨ Bundesanstalt fur Gewasserkunde, Am Mainzer Tor 1, D-56068 Koblenz, Germany c ¨ ¨ ECT Oekotoxikologie GmbH, Bottgerstrasse 2-14, D-65439 Florsheim am Main, Germany d ¨ ¨ Received 15 October 2003; received in revised form 12 February 2004; accepted 13 February 2004 Abstract The leaching behaviour of six selected pharmaceuticals was tested in different soils. Leaching experiments are a part of environmental risk assessment to estimate the distribution and fate of these pharmaceuticals in the environment. Based on the results of this assessment their mobility in soil and their potential to contaminate groundwater was evaluated. When assessing the leaching behaviour of these compounds, the influence of the properties (e.g. grain size distribution, pH, C ) of different soils has to be taken into account. The test results indicated that the leaching org potential found could be rated as low for diazepam, ibuprofen, ivermectin and carbamazepine. Therefore, contamination of the groundwater with these substances seems to be unlikely if the groundwater level is covered with sufficient layers of the soils investigated. Clofibric acid and iopromide were very mobile under the experimental conditions and thus, groundwater contamination would be possible if the soil is exposed to these pharmaceuticals, i.e. wastewater irrigation. These results are more or less in agreement with groundwater monitoring data found in the literature for ibuprofen and diazepam which were in general not present in groundwater, while clofibric acid and iopromide were frequently detected. However, a discrepancy was found for carbamazepine, since it occurs very often in groundwater. This discrepancy might be explained by the fact that the leaching tests were performed with soil, whereas in reality the groundwater contamination occurs mainly over river sediments and sub soil from receiving waters. 2004 Elsevier B.V. All rights reserved. Keywords: Environmental risk assessment; Pharmaceuticals; Leaching; Groundwater contamination 1. Introduction To date, the behaviour and the effects of phar- maceuticals in the environment are widely *Corresponding author. Tel.: q49-6145-9564-50; fax: q49- 6145-9564-99. E-mail address: [email protected] (J. Rombke). ¨ unknown. Most studies performed are restricted to the detection of pharmaceuticals in aqueous matri- ces. The ubiquitous occurrence of human phar- maceuticals in the environment underscores that an environmental risk assessment (ERA) for these substances is essential (Daughton and Ternes, 1999; Sacher et al., 2001; Heberer, 2002). In

Upload: j-oppel

Post on 12-Sep-2016

212 views

Category:

Documents


0 download

TRANSCRIPT

Science of the Total Environment 328(2004) 265–273

0048-9697/04/$ - see front matter� 2004 Elsevier B.V. All rights reserved.doi:10.1016/j.scitotenv.2004.02.004

Leaching behaviour of pharmaceuticals in soil-testing-systems: apart of an environmental risk assessment for groundwater

protection

J. Oppel , G. Broll , D. Loffler , M. Meller , J. Rombke *, Th. Ternesa b c d d, c¨ ¨

ENVIRON Germany GmbH, Muhlwiese 9, D-65779 Kelkheim, Germanya ¨Hochschule Vechta, Abt. F. Geo- und Agrarokologie (ISPA), Postfach 1553, D-49364 Vechta, Germanyb ¨

Bundesanstalt fur Gewasserkunde, Am Mainzer Tor 1, D-56068 Koblenz, Germanyc ¨ ¨ECT Oekotoxikologie GmbH, Bottgerstrasse 2-14, D-65439 Florsheim am Main, Germanyd ¨ ¨

Received 15 October 2003; received in revised form 12 February 2004; accepted 13 February 2004

Abstract

The leaching behaviour of six selected pharmaceuticals was tested in different soils. Leaching experiments are apart of environmental risk assessment to estimate the distribution and fate of these pharmaceuticals in the environment.Based on the results of this assessment their mobility in soil and their potential to contaminate groundwater wasevaluated. When assessing the leaching behaviour of these compounds, the influence of the properties(e.g. grain sizedistribution, pH, C ) of different soils has to be taken into account. The test results indicated that the leachingorg

potential found could be rated as low for diazepam, ibuprofen, ivermectin and carbamazepine. Therefore, contaminationof the groundwater with these substances seems to be unlikely if the groundwater level is covered with sufficientlayers of the soils investigated. Clofibric acid and iopromide were very mobile under the experimental conditions andthus, groundwater contamination would be possible if the soil is exposed to these pharmaceuticals, i.e. wastewaterirrigation. These results are more or less in agreement with groundwater monitoring data found in the literature foribuprofen and diazepam which were in general not present in groundwater, while clofibric acid and iopromide werefrequently detected. However, a discrepancy was found for carbamazepine, since it occurs very often in groundwater.This discrepancy might be explained by the fact that the leaching tests were performed with soil, whereas in realitythe groundwater contamination occurs mainly over river sediments and sub soil from receiving waters.� 2004 Elsevier B.V. All rights reserved.

Keywords: Environmental risk assessment; Pharmaceuticals; Leaching; Groundwater contamination

1. Introduction

To date, the behaviour and the effects of phar-maceuticals in the environment are widely

*Corresponding author. Tel.:q49-6145-9564-50; fax:q49-6145-9564-99.

E-mail address: [email protected](J. Rombke).¨

unknown. Most studies performed are restricted tothe detection of pharmaceuticals in aqueous matri-ces. The ubiquitous occurrence of human phar-maceuticals in the environment underscores thatan environmental risk assessment(ERA) for thesesubstances is essential(Daughton and Ternes,1999; Sacher et al., 2001; Heberer, 2002). In

266 J. Oppel et al. / Science of the Total Environment 328 (2004) 265–273

general, soil can be exposed to human pharmaceu-ticals via the following scenarios:(a) using digest-ed sewage sludge as fertilizer on agricultural fields;(b) irrigation of treated wastewater on fields forwatering of crops or groundwater recharge;(c)leakages of sewer drains and of sewage treatmentplants;(d) flooding of fields with receiving waterscontaining appreciable proportions of treated was-tewater. Veterinary pharmaceuticals such as iver-mectin enter soils mainly by using liquid manureor dung of medicated animals as fertilizers ordirectly during medical treatment of cattle in pas-tures. However, it has to be pointed out that thesix pharmaceuticals selected as test substanceswere chosen as examples, in order to study theleaching potential of such chemicals. Selectioncriteria were their occurrence in groundwater andtheir different medical indication groups as wellas their physico–chemical properties.

Within the last 10 years, several concepts foran environmental risk assessment of pharmaceuti-cals have been proposed(e.g. Stuer-Lauridsen etal., 2000; Rombke et al., 2001a,b). However, only¨for veterinary pharmaceuticals an EU guidelinecovering aquatic as well as terrestrial scenarios isavailable(EU, 1997).

The aim of this work was to investigate theleaching behaviour of six selected pharmaceuticalsin different soils. The compounds selected werechosen either due to their high leaching potential,their intensive application or due to their(eco)toxicological potency. In order to reach this aim,laboratory tests according to internationally accept-ed guidelines had to be performed. The results ofthese tests were used to assess the potential forgroundwater contamination by the selected phar-maceuticals. This contribution is considered toimprove the understanding of the distribution, fateand impact of human and veterinary pharmaceuti-cals in the environment.

2. Materials and methods

2.1. Chemicals

The antiepileptic carbamazepine(CAS No. 298-46-4), the lipid regulator clofibric acid(CAS No.882-09-7), the tranquilizer diazepam(CAS No.

439-14-5), the antiphlogistic ibuprofen(CAS No.155687-27-1) and the parasiticide ivermectin(CAS No. 70288-86-7) were purchased from Sig-ma, Germany. The X-ray contrast medium ioprom-ide (CAS No. 73334-07-3) and its derivatives5-amino-2,4,6-triiodo-isophthalic acid(ATI, CASNo. 35453-19-1), desmethoxyacetyl-iopromide(DAMI ) and (N-2,3-dihydroxypropyl)-5-amino-2,4,6-triiodo-isophthalic acid amide(ATH, CASNo. 111453-32-8) were a courtesy from ScheringAG, Berlin, Germany. Additionally, two radio-labeled compounds were applied:w2- Cx-diaze-14

pam (specific activity: 7.22 MBqymg; AmershamPharmacia Biotech UK Limited, UK) and wring-

Cx-iopromide (specific activity: 1.86 MBqymg;14

courtesy of Schering AG, Berlin, Germany). Thechemical structures, CAS registry numbers and theuseyorigin of these compounds are shown in Table1.

2.2. Soils

The soils selected for the leaching studies cov-ered a wide range of the soil spectrum with respectto organic content and pH. Deviating from OECDguideline (OECD, 2003) only two different soilswstandard soil LUFA 2.2(LU), Euro Soil No. 5(E5)x were used in the experiments with non-labeled compounds, whereas in the experimentsusing C-labeled iopromide and C-labeled diaz-14 14

epam a third soil was investigatedwNeuenkirchen(NK)x. Data on the soil characteristics are givenin Table 2. Leachates were not characterised sincethis measurement is not required by the OECDguideline.

2.3. Experimental set-up of the soil column leach-ing tests

All tests were performed according to the OECDguideline Leaching in Soil Columns,(OECD,2003). The air-dried and sieved soils(-2 mmmesh) were packed in sectionable glass columnsto a height of approximately 30 cm. To obtainuniform packing, the soil was added in smallportions under gentle vibration of the column.Subsequently, the soils were saturated with artifi-cial rain (0.01 molyl CaCl ) to their maximal2

267J. Oppel et al. / Science of the Total Environment 328 (2004) 265–273

268 J. Oppel et al. / Science of the Total Environment 328 (2004) 265–273

Table 2Soils applied in leaching experiments

Name LUFA 2.2 (LU) Euro Soil 5(E5) Neuenkirchen(NK)

pH (CaCl )2 5.8 2.9 7.0C w%xorg 2.3 6.3 1.3Org. matterw%x 4.0 10.8 2.2

WHC w%x 51 25d 33Soil texture loamy sanda loamy sandd Silty loamc

Clay w%x 8.2a 8.4d 17.0c

Silt w%x 17.0a 17.7d 78.4c

Sandw%x 74.8a 73.9d 4.7c

Soil type Gleysolb Podzol Luvisolc

Sampling depth 0–10 cma 0–10 cm 0–10 cmc

Sampling horizon Aah Ah Acp

Land Use hay meadowa pine forest Winter wheatc

Sampling date 07y10y2001a 07y12y2001 08y10y2001

CECscation exchange capacity; WHCswater holding capacity; data based on standard certificate of analysis, Landwirtschaf-a

tliche Untersuchungs-und Forschungsanstalt Speyer, Germany; pers. comm. Weller(Landwirtschaftliche Untersuchungs-und For-b

schungsanstalt Speyer, Germany); data provided by Institut fur Geookologie, Technische Universitat Braunschweig, Germany;c ¨ ¨¨data provided by Geocomp, Bad Vilbel, Germany.d

water holding capacity according to OECD. Thetest substances were applied on the top of the soilcolumns as aqueous solutions or dissolved inorganic solvent at a concentration level of 100mgykg soil (dry weight). All tests were performedin the darkness at a temperature of 20"2 8C. Atotal amount of 393 ml artificial rain(0.01 molylCaCl ) corresponding to 200 mm rainfall was2

added drop wise within 48 h on each soil column.Glass sinter disks on top of the columns ensuredan even distribution of the artificial rain.

The volumes of the leachates collected duringthe test period varied between 361.2 and 417.2 ml.At the end of the experiments, the quantities ofthe non-labeled pharmaceuticals, respectively, thetotal radioactivity contained in the leachates weredetermined. Additionally, in the studies using C-14

labeled substances, the total radioactivity in differ-ent layers of the soil columns was measured. Nofurther measurements(e.g. the pore volume) weredone since such activities are not required by theOECD guideline.

2.4. Chemical analysis

2.4.1. Non-labeled pharmaceuticalsThe non- C labeled compounds in the leachates14

were analysed via solid phase extraction and GC-

MS and LC-tandem MS detection. Further detailsof the methods for the acidic pharmaceuticals(ibuprofen, clofibric acid) are reported in Stumpfet al. (1999), for carbamazepine and diazepam inTernes et al.(2001) and for iopromide in Ternesand Hirsch(2000). For ivermectin a new methodwas developed, which is described in Loffler and¨Ternes(2003).

2.4.2. C-labeled pharmaceuticals14

In experiments with C-labeled substances the14

total radioactivity contained in the water phasewas determined by liquid scintillation counting(LSC), measuring water samples mixed with LSCcocktail (Ultima Gold F, Canberra Packard, Ger-many) in a LSC counter (TriCarb 2500 TR,Canberra Packard, Dreieich, Germany). The totalradioactivity in the soil samples was quantified bycombustion of soil aliquots in a sample oxidiser(TriCarb 307, Canberra Packard, Dreieich, Ger-many) and LSC counting of the resulting trappingagent-LSC cocktail mixture(Carbosorb and Per-mafluor, Canberra Packard, Dreieich, Germany).An additional chemical analysis was conductedwith leachates from experiments with non-labeledand C-labeled iopromide. For this, leachate sam-14

ples of 15 ml were mixed with 175 ml acetonitrile.The liquid phase was then evaporated to dryness

269J. Oppel et al. / Science of the Total Environment 328 (2004) 265–273

Table 3Recovery in % of applied non-labeled pharmaceuticals in the leachates(means of two soil columns and absolute deviation of mean)

Carbamazepine Clofibric acid Diazepam Ibuprofen Iopromide Ivermectin

LU E5 LU E5 LU E5 LU E5 LU E5 LU E5

Recoveryw%x 0"0 0"0 61"9 4"1 0"0 0"0 0"0 0"0 0"0 38"7 0"0 0"0

Fig. 1. Leaching of C-diazepam in LU, E5 and NK. Recovery14

rates in% of applied test substance in soil and leachates(meansof measured soil aliquots of two columns(ns8); error bars:S.D.; n.d.: recovery-0.5%).

Fig. 2. Leaching of C-iopromide in LU, E5 and NK. Recov-14

ery rates in % of applied test substance in soil and leachates(means of measured soil aliquots of two columns(ns8); errorbars: S.D.; n.d.: recovery-0.5%).

by a rotary evaporator at 408C and 150 mbar.Then further TLC analyses were conducted, fol-lowing procedures described by Kalsch(1992).

3. Results

3.1. Non-labeled pharmaceuticals

The majority of the pharmaceuticals tested wereretained in the two soils investigated. Carbamaze-pine, diazepam, ibuprofen and ivermectin were notdetected in the leachates of the soils LU and E5(Table 3), neither were their metabolites in humans10,11-dihydro-10,11-dihydroxycarbamazepine,oxazepam nor 2-hydroxy-ibuprofen. In contrast,the LU leachate contained 61% of the clofibricacid applied, while only 4% were detected in theleachate of E5. Iopromide was not detected in theleachates of LU, whereas 38% of the appliediopromide were detected in the leachate of the soilE5.

3.2. C-labeled compounds14

In the experiments with C-labeled diazepam14

no radioactivity could be detected in any soilleachates(LU, E5, NK). This is in good accor-dance with the experiments using non-labeled diaz-epam. Hence, neither diazepam nor any TPs ofdiazepam leached through the soil columns. Theradioactivity initially present as diazepam waslocated in the soil E5 exclusively in the uppermostsoil layer (0–5 cm), whereas in the soils LU andNK radioactivity was found down to a soil depthof 15 cm and 20 cm, respectively. Below thesedepths only negligible amounts of radioactivity(-0.5%) were detected(Fig. 1).

All leachates in the experiments with C-14

iopromide contained substantial amounts of radio-activity, LU (78%), E5 (59%) and NK (47%).

270 J. Oppel et al. / Science of the Total Environment 328 (2004) 265–273

The vertical extension of the radioactivity in thesoil column, corresponding to C-iopromide and14

transformation products(TPs) formed, was slightlyhigher in LU than in E5 and in NK(Fig. 2).

4. Discussion

4.1. Carbamazepine

The mobility of carbamazepine(pK s7,a

log K s3.5 Jones et al., 2002) did not differ inOW

the two soils investigated. Obviously, the retentionof carbamazepine was not decisively affected bythe distinctly differing properties of both soils,concerning pH and C . In contrast, Drewes et al.org

(2002) observed no attenuation of carbamazepinein subsoil during bank filtration. Therefore, it canbe assumed that either the water volume appliedwas too low for a comparable leaching of carba-mazepine or that the usage of topsoil resulted ineffects, such as an increased sorption andyor biod-egradation. In order to elucidate these assumptions,it would be helpful to repeat the experiments withradiolabeled carbamazepine, which was not avail-able in the study performed, including a chemicalanalysis of the radioactivity in soils and leachates.However, carbamazepine is present in groundwatervery often (Sacher et al., 2001; Heberer, 2002).This might be explained by the fact that ground-water contamination occurs mainly over river sed-iments and subsoil from receiving waters(Ternes,2001; Mersmann et al., 2002). Since topsoil andsubsoil may differ significantly in C , bacterialorg

community, and further properties, the results fromleaching tests performed with topsoil have only alimited transferability for subsoil or aquiferscenarios.

4.2. Clofibric acid

The sorption behaviour and thus, the mobilityof organic acids(e.g. salicylic acid) in soils isstrongly pH dependent(Dubus et al., 2001).Organic acids are widely present in their neutral,undissociated form at pH conditions below theirpK -value (Schwarzenbach et al., 2003). Thisa

results in a distinctly higher lipophilicity andusually in a higher tendency to sorb onto organic

soil matter for the undissociated acids comparedwith their more polar dissociated form.

It can be assumed that the relatively low pH of2.9 in the soil E5 led to an appreciable higherretention of clofibric acid(estimated pK of 2.84,a

log K s2.57 in Hansch et al., 1990), comparedOW

to the soil LU, due to enhanced sorption of theundissociated acid onto soil particles. In soil LUwith a more common pH of 5.8 a significantlyhigher mobility of clofibric acid was observed.This is in good agreement with findings of Scheyttet al. (2001) and Heberer(2002) who reported analmost tracer—like movement of clofibric acid insoil columns and during riverbank filtration.

Although a transformation of clofibric acid can-not entirely be excluded, it is very unlikely sinceclofibric acid remains persistent during wastewatertreatment, riverbank filtration and under otherenvironmental conditions as reported by variousauthors(Stumpf et al., 1999; Winkler et al., 2001;Heberer et al., 2002; Weigel et al., 2002). It canbe concluded that groundwater contamination dueto the infiltration of clofibric acid through topsoilis possible.

4.3. Diazepam

Diazepam is a lipophilic substance with alog K of 2.82 and a pK of 3.3 (Stuer-LauridsenOW a

et al., 2000) and showed a very low mobility inall soils. Even in the relatively acidic soil E5 asignificantly increased mobility due to a protonatedamino moiety was not observed. It can be expectedthat its leaching behaviour was mainly determinedby the organic carbon content of the soils. Anextensive transformation of diazepam in the soil isunlikely, since diazepam was widely stable in awaterysediment test under aerobic conditions(Lof-¨fler, 2003) and transformation products might haveshown certain mobility in the soil due theirincreased polarity. Therefore, the leaching potentialfor diazepam should be estimated as very low.

4.4. Ibuprofen

The weak organic acid ibuprofen contains acarboxylic moiety with a pK of 4.9(Merck, 2001)a

and a distinct lipophilicity in its undissociated

271J. Oppel et al. / Science of the Total Environment 328 (2004) 265–273

Table 4Composition of radioactivity in leachates(two replicates;errorsabsolute deviation of mean) of the tests performed withiopromide; the TPs abbreviations are given in Section 2.1

Compound R -valuef Soil LU Soil E5

Unknown TP 1 0.20 63"2 15"9ATI 0.22 – –Unknown TP 2 0.31 23"1 –ATH 0.32 – –IO 0.37 – 85"9DAMI 0.47 – –Unknown TP 3 0.70 14"3 –

Table 5Percentage of initially applied C-labeled and non-labeled14

iopromide recovered unchanged in soil leachates(2 replicates;errorsabsolute deviation of mean)

Soil LU 2.2 EuroSoilE5

C-labeled iopromide14 0% 50"1%non-labeled iopromide 0% 38"7%

form (log K of 3.5 Stuer-Lauridsen et al., 2000;OW

Jones et al., 2002). In both tested soils with a pHof 5.8 (LU) and 2.9(E5), the carboxyl group ofibuprofen should be at least partly protonatedleading to a sorption onto soil particles. Further-more, it is known that ibuprofen can be trans-formed easily under environmental conditions(Ternes, 1998; Winkler et al., 2001). Since ibupro-fen was not found in the soil leachates, it can beexpected that the soil passage mean an effectivebarrier for ibuprofen, due to sorption andyor bymicrobial degradation.

4.5. Iopromide

Iopromide is a polar compound(log K of yOW

2.33) with a high water solubility(Steger-Hart-mann et al., 1999). Its tendency for binding to soilparticles can be expected to be very low, sincespecific interactions with soil particles areunknown for iopromide. Due to its high polarity,iopromide leaches rapidly through the soil, nearlyunimpeded by interactions with soil particles.

In experiments with C-labeled iopromide,14

more than 50% of the radioactivity applied wasfound in the leachates of both soils. Contrary non-labeled iopromide was not found in leachates ofsoil LU and leachates of soil E5 contained only38"7% of the applied iopromide.

In order to clarify the virtually contradictoryresults, leachate samples of the experiments with

C-labeled iopromide were analysed by radio14

TLC, applying several potential environmentaltransformation products(TPs) as reference com-pounds. TLC analysis showed that iopromide was

completely transformed in soil LU, under forma-tion of at least three TPs(Table 4).

The R -values of the TPs 2 and 3 almostf

matched those of the reference compounds ATIand ATH. In order to determine, whether ATI orATH were formed as TPs, additional LC-tandemMS experiments were conducted exhibiting thatthe presence of the iopromide derivatives ATI,ATH or DAMI can be excluded. During passageof the soil E5 iopromide was transformed only toa minor extent, while 85"9% of the iopromideremained unchanged. The formed product is pre-sumably the same compound as the unknown TP1, since theirR -values matched very well.f

Finally, the results of the experiments with C-14

labeled and non-labeled iopromide were linked inTable 5, to investigate the comparability of theexperiments. In both experiments with soil LUiopromide was never found in the leachates. Theiopromide content in leachates of soil E5 also didnot differ significantly in experiments with radiol-abeled and non-labeled iopromide. This showsclearly that both analytical methods led reprodu-cibly to the same results.

It might be assumed that the varying mobilityof the iopromide equivalents in the three soils wasmainly a result of the differing extent oftransformation.

Hence, the at least partial transformation ofiopromide during soil passage gives some evidenceto the idea, that the removal of iopromide duringriverbank filtration is caused by its degradation(Putschew et al., 2000).

It can be concluded that the introduction ofiopromide onto soils means a certain risk forgroundwater contamination, since in any case agroundwater contamination with iopromide or itsTPs is very likely.

272 J. Oppel et al. / Science of the Total Environment 328 (2004) 265–273

4.6. Ivermectin

Concluding from the experimental data, iver-mectin shows no leaching potential in soil col-umns. Comparable data for ivermectin werereported by Halley et al.(1989a), and for theivermectin derivative abamectin by Gruber et al.(1990) and by Bull et al.(1984). The immobilityof the test compound in the soils tested and itstendency to sorb onto soil particles is related to itshigh lipophilicity (log K s3.22, K s12 660–OW OC

15 700 Halley et al., 1989b).However, Tolls(2001) could not assign the soil

affinity of avermectins exclusively to their lipo-philicity. LC-MS data shows that ivermectin iscapable to form adducts with cations, such asammonium or sodium(Ali et al., 2000). Hence,there is some reason for the presumption of aspecific binding of ivermectin to soil, which mightdeviate from the formation of adducts or complex-es with immobile inorganic soil matter.

5. Conclusion

Based on the results of the experiments, it canbe presumed that carbamazepine, diazepam, ibu-profen and ivermectin have to be classified as non-mobile in the soils tested. That means that thesechemicals, once applied onto the soil surface, willremain in the upper soil layers and may either bebound by soil particles or may be transformed inthe soil. However, a discrepancy was found forcarbamazepine, since it occurs very often ingroundwater. This might be explained by the factthat the leaching tests were performed with soil,whereas in reality the groundwater contaminationoccurs mainly over river sediments and sub soilfrom receiving waters.

However, the observed mobility of clofibric acidand iopromide in the soil columns indicates apotential risk for groundwater contamination.These results are in general agreement with theoccurrence of both substances in the groundwaterof several regions in Germany, since the ground-water contamination detected was mainly causedby an infiltration of these pharmaceuticals fromriver water through sediment and subsoil into thegroundwater. Pharmaceutical substances, which

enter the aquifer may pose a general problem,especially if they are persistent. The presence ofthese chemicals in the groundwater might lead todrinking water contamination and ecotoxicologicalrisks. Due to the low microbial activity in thedeeper soil layers these contaminants may bepersistent in the aquifer over decades and may,therefore cause a long-term risk for the ground-water community.

Especially the compartment ground water is asubject of high interest not only from an ecotoxi-cological point of view but also due to toxicolog-ical reasons. Therefore, the assessment of potentialrisks of chemicals to human health via ground-water contamination must always be considered.Based on the results of this project the next stephas to be the development of useful test-systemsin order to determine the toxicity of selectedcontaminants to groundwater organisms. Depend-ing on their ecotoxicological effects threshold lev-els especially for pharmaceuticals, which have apotential to leach into the ground water should bedefined in the near future.

Acknowledgments

Financial support was kindly granted by theGerman Federal Environmental Agency(RqD-No. 299 67 401). The authors wish to thank theSchering A.G., Germany for providing severalanalytes and PD Dr R. Kreuzig, T.U. Braun-schweig, Germany for providing the soil NK.

References

Ali SM, Sun T, McLeroy GE, Philuppo ET. Confirmation ofeprinomectin, moxidectin, abamectin, doramectin, and iver-mectin in beef liver by liquid chromatographyypositive ionatmospheric pressure chemical lonization mass spectrometry.J AOAC Int 2000;83:39–52.

Bull DL, Ivie GW, MacConnel JG, Gruber VF, Ku CC, ArisonBH, Stevenson JM, VandenHeuvel WJA. Fate of AvermectinB1a in soil and plants. J Agri Food Chem 1984;32:94–102.

Daughton CG, Ternes TA. Pharmaceuticals and personal careproducts in the environment: agents of subtle change?Environ Health perspect 1999;107(Suppl. 6):907–938.

Drewes JE, Heberer T, Reddersen K. Fate of pharmaceuticalsduring indirect potable reuse. Water Sci Technol2002;46:73–80.

273J. Oppel et al. / Science of the Total Environment 328 (2004) 265–273

Dubus IG, Barriuso E, Calvet R. Sorption of weak organicacids in soils: clofencet, 2,4-D and salicylic acid. Chemo-sphere 2001;45:767–774.

EU. Note for Guidance: Environmental Risk Assessment forVeterinary Medicinal Products other than GMO-Containingand Immunological Products. EMEAyCVMPy055y96. Eur-opean Union, EMEA, London(1997).

Gruber VF, Halley BA, Hwang S-C, Ku CC. Mobility ofavermectin B1a in soil. J Agric Food Chem 1990;38:886–890.

Halley BA, Jacob TA, Lu AYH. The environmental impact ofthe use of ivermectin: environmental effects and fate. Che-mosphere 1989;18:1543–1563.

Halley BA, Nessel RJ, Lu AYH, Roncalli RA. The environ-mental safety of ivermectin: an overview. Chemosphere1989;18:1565–1572.

Hansch C, Sammes PG, Taylor JB. Comprehensive medicinalchemistry, the rational design, mechanistic study and thera-beutic application of chemical compounds. Cumulative sub-ject index and drug compendium. Oxford: Pergamon, 1990.

Heberer T. Occurrence, fate, and removal of pharmaceuticalresidues in the aquatic environment; a review. Toxicol Lett2002;131:5–17.

Heberer T, Reddersen K, Mechlinski A. From municipalsewage to drinking water: fate and removal of pharmaceu-tical residues in the aquatic environment in urban areas.Water Sci Technol 2002;46:81–88.

Jones OAH, Voulvoulis N, Lester JN. Aquatic environmentalassessment of the top 25 English prescription pharmaceuti-cals. Water Res 2002;36:5013–5022.

Kalsch W, Abbau iodhaltiger Rontgenkontrastmittel in der¨Umwelt, Dissertation:Johannes Gutenberg Universitat Mainz¨(1992).

Loffler D. Residues of human and veterinary pharmaceuticals¨in aquatic compartments: Analysis, Distribution and Fate inWater and Sediment, Dissertation, Johannes GutenbergUniversitat in Mainz, 2003.¨

Loffler D, Ternes TA. Analysis of acidic pharmaceuticals,¨antibiotics and ivermectin in river sediment using LC-tandem MS. J Chromatogr A 2003;2021:133–144.

Merck. The merck index. London: Merck Publications, 2001.Mersmann P, Scheytt T, Heberer T. Column experiments on

the transport behavior of pharmaceutically active compoundsin the saturated zone. Acta Hydrochim Et Hydrobiol2002;30:275–284.

OECD. Guideline for Testing of Chemicals. Leaching in SoilColumns. In: Guideline for Testing of Chemicals. Leachingin Soil Columns. Paris(2003).

Putschew A, Wischnack S, Jekel M. Occurrence of triiodinatedX-ray contrast agents in the aquatic environment. Sci TotalEnviron 2000;255:129–134.

Rombke J, Knacker T, Teichmann H. Ecotoxicological evalu-¨ation of pharmaceuticals. In: Kummerer K, editor. Pharma-¨

ceuticals in the environment. Berlin–Heidelberg–New York:Springer-Verlag, 2001a. p. 123–141.

Rombke J, Knacker T, Teichmann H. Environmental risk¨assessment of pharmaceuticals: a proposal with specialemphasis on european aspects. In: Daughton CG, Jones-Lepp T, editors. Pharmaceuticals and personal care productsin the environment: scientific and regulatory issues. Wash-ington, DC: American Chemical Society, 2001b. p. 304–319.

Sacher F, Lange FT, Brauch H-J, Blankenhorn I. Pharmaceu-ticals in groundwaters. Analytical methods and results of amonitoring program in Baden-Wurttemberg, Germany. J¨Chromatogr A 2001;938:199–210.

Scheytt T, Mersmann P, Heberer T. Natural Attenation ofPharmaceuticals. In: 2nd International Conference on Phar-maceuticals and Endocrine disrupting Chemicals in Water,Minneapolis, Minnesota,(2001); 253–259.

Schwarzenbach RP, Gschwend PM, Imboden DM. Environ-mental organic chemistry. New York, Chichester, Brisbane,Toronto, Singapore: John Wiley and Sons, 2003.

Steger-Hartmann T, Lange R, Schweinfurth H. Environmental¨risk assessment for the widely used iodinated X-ray contrastagent iopromide (Ultravist). Ecotoxicol Environ Saf1999;42:274–281.

Stuer-Lauridsen F, Birkved M, Hansen LP, Holten Lutzhoft H-¨C, Halling-Sorensen B. Environmental risk assessment ofhuman pharmaceuticals in Denmark after normal therapeuticuse. Chemosphere 2000;40:783–793.

Stumpf M, Ternes TA, Wilken R-D, Rodrigues SV, BaumannW. Polar drug residues in sewage and natural waters in thestate of Rio de Janeiro, Brazil. Sci Total Environ1999;225:135–141.

Ternes TA. Occurrence of drugs in German sewage treatmentplants and rivers. Water Res 1998;32:3245–3260.

Ternes TA. Pharmaceuticals and metabolites as contaminantsof the aquatic environment. In: Daughton CG, Jones-LeppTL, editors. Pharmaceuticals and personal care products inthe environment. Washington, DC: American ChemicalSociety, 2001.

Ternes TA, Hirsch R. Occurence and behaviour of X-raycontrast media in sewage facilities and the aquatic environ-ment. Environ Sci Technol 2000;34:2741–2748.

Ternes TA, Bonerz M, Schmidt T. Determination of neutralpharmaceuticals in wastewater and rivers by liquid chro-matography-electrospray tandem mass spectrometry. J Chro-matogr A 2001;938:175–185.

Tolls J. Sorption of veterinary pharmaceuticals in soils: areview. Environ Sci Technol 2001;35:3397–3406.

Weigel S, Kuhlmann J, Huhnerfuss H. Drugs and personal¨care products as ubiquitous pollutants: occurence and distri-bution of clofibric acid, caffeine and DEET in the NorthSea. Sci Total Environ 2002;295:131–141.

Winkler M, Lawrence JR, Neu TR. Selective degradation ofibuprofen and clofibric acid in two model river biofilmsystems. Water Res 2001;35:3197–3205.