Determination of non-steroidal anti-inflammatory drug (NSAIDs) residues in water samples

https://doi.org/10.1016/j.envint.2004.12.001Get rights and content

Abstract

Pharmacologically active substances used to treat human and animal illnesses can enter the aquatic environment via effluents from wastewater treatment plants or in the case of veterinary drugs directly through liquid manure discharge. Some of these substances enter the environment either as the parent compound or as active/inactive metabolites. Due to their pharmacological activity, their determination and understanding their behavior and fate in the environment are important.

The scope of this paper was to develop an analytical procedure to determine common pharmaceutical residues in wastewaters. Pharmacologically active substances were chosen according to their wide spread application in Slovenia and Central Europe and are members of analgesics, e.g., non-steroidal anti-inflammatory drugs: ibuprofen, naproxen, ketoprofen and diclofenac. Selected compounds were isolated from synthetic water using a novel SPE sorbent Strata™ X. Due to the non-volatile nature of these compounds they were first silylised prior to gas chromatographic-mass spectrometric detection. The developed procedure was tested with synthetic wastewaters and their extraction efficiency (>84%) and method limits of detection (2–6 ng L−1) were determined. Our procedure has been adopted and optimised for “real” water samples and applied to eleven drinking and ten river water samples from Slovenia. The results showed no traces of NSAIDs in all potable water samples and low-range contamination (ng L−1) of Slovene rivers. These results show that NSAIDs contamination of Slovene waters is comparable with published results of water contamination in Central Europe.

Introduction

In recent years increasing attention has been directed toward the discharge, presence and potential effects of pharmaceuticals in the environment. Thousands of tons of pharmacologically active substances are used yearly to treat or prevent illnesses, or to help people face the stresses of modern life. The discharge of therapeutic agents from production facilities, hospitals and private household effluent as well as improper disposal of unused drugs pose a burden on the environment (Christensen, 1998). Pharmaceuticals are released into the environment either as the parent compound or as active/inactive metabolites. Thus, often it is not only the parent compound which should be the subject for a risk assessment but also the active metabolites (Christensen, 1998, Halling-Sørensen et al., 1998). Concentrations of pharmaceutical residues measured in water may give rise to human exposure in the ng per day range, which is at least three to four orders of magnitude lower than that required to produce a pharmacological effect. Risks arising from acute exposure can therefore be regarded as unlikely. However, possible effects of life-long exposures have still to be determined (Christensen, 1998).

Pharmaceuticals have been selected or designed due to and because of their biological activity. In respect to their purpose they should be considered as suspicious environmental contaminants (Christensen, 1998). Furthermore, they often have low biodegradability, and can accumulate, reaching detectable and biologically active amounts (Zuccato et al., 2000). Quantitative evaluation of the fate of pharmaceuticals in the aquatic environment, proper risk assessment and improvement of the efficiency of sewage treatment plants need sensitive and reliable analytical methods.

There are no data regarding pollution with pharmaceutical residues in Slovenia. Therefore, the aim of our study was to develop an analytical procedure, which allows the quantification of pharmaceuticals in water at the ng L−1 level. By analysing tap, well and river samples from around Slovenia, we hope to gauge the extent of pharmaceutical residues in Slovene waters. Model compounds were selected among the pharmaceuticals, which predominate in the analyses of environmental samples, as well as on the lists compiled from prescription data. Most of these pharmaceuticals belong to the class of analgesics (non-steroidal anti-inflammatory drugs, NSAIDs), antibiotics, antihypertensives, antiasthmatics, diuretics and psycholeptics (Kümmerer, 2001a). For this reason, the following four pharmaceuticals from the class of NSAIDs were chosen as model compounds: ibuprofen, naproxen, diclofenac and ketoprofen. The four investigated drugs belong to a group of the most commonly prescribed drugs between non-steroidal anti-inflammatory drugs. Data from annual reports (Oražem and Pečar-Čad, 2000, Oražem and Pečar-Čad, 2001, Oražem and Pečar-Čad, 2002) show a quantity of the drugs dispensed by prescriptions from health-centres. However, these data underestimate the total use of pharmaceuticals in Slovenia, because drugs dispensed over-the-counter and those spent in hospitals also contribute to the total. The quantities of annually prescribed pharmaceuticals have been published (Oražem and Pečar-Čad, 2000, Oražem and Pečar-Čad, 2001, Oražem and Pečar-Čad, 2002). The quantity of naproxen, together with other NSAID representatives, is the most outstanding in the group of investigated drugs and is reported to be between 1.9 and 2.6 tons/year. Furthermore, naproxen is eliminated partly unmetabolised (60%) and is persistent in the environment. Naproxen is therefore expected to pose the biggest load (among the four investigated drugs) on the Slovenian aquatic environment.

Pharmaceutical residues are usually present in environmental water samples in trace levels. The most common sample isolation and pre-concentration technique is solid-phase extraction (SPE) (Rodríguez et al., 2003) where as well as isolation and pre-concentration, the matrix-solvent (water) is exchanged with a more volatile organic solvent suitable for gas chromatography (GC). Due to low vapour pressure, gas chromatographic separations of selected NSAIDs can be performed only after derivatisation of the native compounds to less polar species (Rodríguez et al., 2003). This involves converting the carboxylic group present on these drugs to the methyl ester derivative using diazomethane (Rodríguez et al., 2003, Öllers et al., 2001, Ternes, 2001, Poole, 1991). The yield of the reaction is usually high, however, because of high toxicity and low stability of diazomethane, alternatives have been proposed. Koutsouba et al. (2003) and Sacher et al. (2001) derivatise the carboxylic group using pentafluorobenzyl bromide with triethylamine as a catalyst. The most widely used alternatives to diazomethane are alkylsilyl reagents (Poole, 1991), namely N-methyl-N-(tert.-butyldimethysilyl) trifluoroacetamide (Rodríguez et al., 2003) or N-methyl-N-(trimethylsilyl) trifluoroacetamide (MSTFA) (Heath, 1998).

An analytical procedure for the determination of NSAIDs in water, based on solid phase extraction (SPE) with a new, patent-pending sample preparation sorbent Strata™ X, followed by derivatisation with MSTFA and GC-MSD analysis was developed and tested on synthetic and authentic well, tap and river water samples.

Section snippets

Chemicals

Sigma-Aldrich Company Ltd (Gillingham, GB) supplied all the drugs under investigation (ibuprofen, diclofenac, naproxen and ketoprofen) and the derivatisation agent MSTFA (N-methyl-N-(trimethylsilyl) trifluoroacetamide). Mecoprop (2-(4-chloro-2-methylphenoxy) propanoic acid) was used as an internal standard and was obtained from Labor. Dr. Ehrenstorfer-Schäfers (Ausburg, Germany). Methanol (MeOH), toluene and 37% hydrochloric acid (HCl) were of analytical grade and were provided by Merck

SPE

Breakthrough of the selected SPE sorbent was investigated using a synthetic wastewater containing approx. 0.1 mg of each of the test compounds and 500 mL of the samples were passed through two cartridges connected sequentially. After the enrichment step, the cartridges were then analysed separately. Since the analytes were not detected in the eluant from the second cartridge, it was proven that the applied cartridge dimension (60 mg/3 mL) was adequate for the quantitative adsorption of the

Conclusions

In the presented work an analytical procedure for determination of pharmaceutical residues in water samples was developed. The qualitative determination of the selected compounds (naproxen, ketoprofen, ibuprofen, diclofenac) included development and optimisation of following analytical steps: SPE, derivatisation and GC-MSD analysis. Under optimal working conditions (flow, solvent volume, cartridge dimension, derivatisation conditions) isolation of selected compounds from water samples with

Acknowledgements

The financial support from the Ministry of Education, Science and Sport (Projects L1-6552 and P1-0143) is acknowledged. The authors are thankful to Mrs. Silva Perko and dr. Hermina Leskovšek for their helpful advice and technical assistance and to all enthusiastic researchers involved in sampling scheme.

Cited by (215)

View all citing articles on Scopus
View full text