Journal of Environmental Monitoring

Determination of six anti-infectives in wastewater using tandem solid-phase extraction and liquid chromatography–tandem mass spectrometry

Pedro A. Segura^a, Araceli García-Ac^a, André Lajeunesse^b, Dipankar Ghosh^c, Christian Gagnon^b and Sébastien Sauvé*^a
aDépartement de Chimie, Université de Montréal, C.P. 6128, succursale Centre-ville, Montréal, QC, Canada H3C 3J7. E-mail: sebastien.sauve@umontreal.ca; Fax: +1 514 343-7586; Tel: +1 514 343-6749
^bAquatic Ecosystem Protection Research Division, Environment Canada, 105, McGill St., Montréal, QC, Canada H2Y 2E7
^cThermo Fisher Scientific, 355 River Oaks Parkway, San Jose, CA, 95143, USA

A rugged and specific method based on tandem solid-phase extraction and liquid chromatography–tandem mass spectrometry for the determination of anti-infectives in raw sewage and wastewater plant effluents was developed. Analyte recoveries from spiked effluents ranged from 68 to 104%. Two specific selected reaction monitoring transitions and their peak area ratios were used to avoid false positives and confirm the presence of the targeted substances. Detection limits allowed low nanogram per litre detection (0.3–22 ng L^–1). The method was successfully applied to real samples from the Montréal wastewater treatment plant. All the studied anti-infectives were found in the wastewater samples in concentrations ranging from 39 to 276 ng L^–1. Mean flows of anti-infectives were estimated from effluent concentrations and it was found that large amounts (>118 g day^–1 up to 830 g day^–1) are discharged in the receiving waters of the St Lawrence River.

Anti-infectives is a general term that refers to several classes of biologically active compounds used to treat or to prevent infections, and includes therapeutic agents of synthetic origin such as the antimicrobials (e.g. sulfonamides, fluoroquinolones) or derived from natural or semi-natural sources such as the antibiotics (e.g. macrolides, tetracyclines). The widespread use of anti-infectives in urban centers, as well as their resistance to biodegradation or elimination in wastewater treatment plants (WWTP), has led to their appearance in WWTP effluents and surface waters.^1–3 In the last few years, there has been a growing concern about the environmental fate and the possible effects of these agents on the aquatic environment.^4–6

The first report on the occurrence of anti-infective traces in the aquatic environment was published as early as 1983.⁷ Later studies^8,9 acknowledged that the source of pharmaceuticals in the environment was mainly the excreta of individuals taking medication at home, in hospitals or clinics, and transported via wastewaters into receiving surface waters. Hence, it is important to know the amounts of these substances released in the aquatic environment to be able to properly evaluate the risks, the effects and the potential impacts of these products.

According to IMS Health Canada,¹⁰ anti-infectives were the 4th most prescribed therapeutic class in 2004 in Canada and almost 5 million such prescriptions were dispensed in Québec (Fig. 1). Since Montréal is the most populous city of the province, significant amounts of anti-infectives in the citys sewage can be expected. However, because Montréals wastewaters are processed by one of the largest primary treatment plants of North America (average flow 2500000 m³ day^–1) the effluent wastewaters are made of a quite complex matrix (Table 1) and a selective extraction method is necessary to properly determine the concentrations of anti-infectives released into the St Lawrence River.

	Fig. 1 Most prescribed anti-infectives in Québec in 2004.

Presently, there are numerous solid-phase extraction (SPE) and liquid chromatrography–mass spectrometry methods published in the literature for the determination of anti-infectives in environmental matrices, but few are able to successfully extract and quantify those compounds in more complex waters. To the authors knowledge, only a handful of published methods show SPE recoveries in raw sewage^11,12 or primary effluents.^13,14 Gobël et al.¹⁴ obtained high recoveries (>80%) of various sulfonamide and macrolide anti-infectives, but lower recoveries of trimethoprim (47% ± 2.8%) in primary-treated effluents. Golet et al.¹³ reported high recoveries (>81%) of different quinolones and fluoroquinolones. However, the later method is based on fluorescence detection and proved to be unsuitable for Montréals wastewaters, which contain important inputs of industrial wastes that seem to interfere with fluorescence detection. When analyzing the Montréal wastewater matrix, Segura et al.¹⁵ observed the presence of intense peaks eluting near the retention times of the analytes, making the quantification of norfloxacin, ciprofloxacin and enrofloxacin impossible by fluorescence. The authors explained the presence of those peaks by the possible presence of organic substances, not found in the Swiss effluent of the original work,¹³ having excitation and emission wavelengths near those of the fluoroquinolones (_ex = 278 nm, _em = 445 nm). Lindberg et al.¹¹ reported good recoveries of sulfamethoxazole (77% ± 5) and trimethroprim (87% ± 9) but weaker recoveries of ciprofloxacin (61% ± 5) in fortified hospital sewage samples, and recently Vieno et al.¹² obtained low absolute recoveries of ciprofloxacin (32% ± 24) in WWTP influents. To properly determine the occurrence of anti-infectives in raw sewage and primary-treated effluents, such as those of the city of Montréal, it is therefore necessary to develop a new rugged and specific method capable to quantify target analytes in spite of high dissolved organic carbon (DOC) concentrations.

The main goals of the present study are: (i) to develop a new SPE method, optimized for the extraction of six highly prescribed (Table 2) anti-infectives from WWTP influents and primary effluents; (ii) to determine the amounts of these compounds in raw sewage and WWTP primary effluents using liquid chromatography–tandem mass spectrometry (LC/MS/MS); (iii) to confirm the presence of the detected compounds using a second specific selected reaction monitoring (SRM) transition and the area ratio of the two transitions.

Anti-infective (acronym)	Structure	MW/g mol–1	pKaa	log Dowb	Internal standard (acronym)	Surrogate standard
Sulfamethoxazole (SMX)		253.2776	1.7,30 1.85,31 5.60,31 5.7032	8833
Trimethoprim (TRI)		290.3177	1.32,34 3.23,31 6.6,35 7.3436	–0.17033
Ciprofloxacin (CIP)		331.3415	5.90,37 6.09,38 8.74,38 8.8937	–1.7833
Levofloxacin (LEV)		361.3675	5.97,37 6.10,39 8.22,38 8.2837	–1.433
Clarithromycin (CLA)		747.9534	8.3,40 8.9941	0.0633
Azithromycin (AZI)		748.9845	8.1,40 8.74,41 8.8,40 9.4541	–0.7733
a Lowest and highest values reported in the literature.  b Estimated values.

Pyrimethamine (PYR), sulfamethoxazole (SMX), diaveridine (DIA), trimethoprim (TRI), ciprofloxacin (CIP), lomefloxacin (LOM), levofloxacin (LEV), azithromycin (AZI), josamycin (JOS), were purchased from Sigma–Aldrich Canada (Oakville, ON, Canada). Clarithromycin (CLA), was kindly provided by Abbot Canada (Montréal, QC, Canada). Disodium ethylenediaminetetraacetate (Na₂EDTA) and formic acid 95% were purchased from Sigma–Aldrich Canada, sodium hydroxide 99% was obtained from Fisher Scientific Canada (Ottawa, ON, Canada) and ammonia 25% was purchased from BDH Chemicals (Toronto, ON, Canada). Solvents used for mobile phase preparation such as water, methanol (MeOH) and acetonitrile (ACN) were LC/MS grade, solvents used for cartridge elution such as MeOH, ACN and isopropanol (i-PrOH) were Optima grade and they were all obtained from Thermo Fisher Scientific. Glass fiber pre-filters (1.2 µm pore size) and MF membranes (0.45 µm pore diameter) were manufactured by Millipore (Billerica, MA, USA). SPE cartridges Strata-X (surface-modified styrene divinylbenzene polymer) 200 mg–6 mL and Strata-X-C (benzenesulfonic acid group bonded on a polymeric surface) 200 mg–3 mL were purchased from Phenomenex (Torrance, CA, USA). Liquid chromatography was carried out using a BetaBasic C18 column (50 × 2.1 mm, 3 µm diameter particle size) manufactured by Thermo Fisher Scientific (Waltham, MA, USA).

SPE work was done on a 12-position vacuum manifold from Phenomenex. SPE eluates were evaporated using a 9-port Reacti-vap unit from Pierce (Rockford, IL, USA). LC/MS/MS analyses were performed on a ThermoFinnnigan HPLC Surveyor System coupled to a TSQ Quantum Ultra triple quadrupole by Thermo Fisher Scientific.

Stock solutions of 200 mg L^–1 were prepared in MeOH, except for the fluoroquinolones (CIP, LOM and LEV) which were dissolved in 5% (0.1 M) NaOH–95% MeOH. All stock solutions were conserved at –15 °C and used for no more than 3 months. ESI source optimization solutions of 1 mg L^–1 of each compound were prepared in 0.1% formic acid in 50% H₂O–25% MeOH–25% ACN. Mixed working solutions containing 500 µg L^–1 of the anti- and infectives, or 500 µg L^–1 of the internal standards, were prepared fresh daily in 0.1% formic acid in 90% H₂O–5% MeOH–5% ACN.

Recovery tests.

A series of 24 h composite samples of the primary effluent was collected in amber bottles at the WWTP. The bottles were placed in a cooler at 4 °C and then transported to the laboratory. They were immediately filtered through 1.2 µm pore size fiber glass filters and then 0.45 µm pore size mixed cellulose ester membranes. Sub samples of 250 mL of the filtered wastewater were then transferred to volumetric glass flasks. To test the recovery, they were combined with a 250 µL aliquot of the 500 µg L^–1 anti-infective mixed working solution. Sample pH was adjusted to 3 using 12.5 mmol of formic acid and 1.0 M NaOH. As a sample additive, 1 mL of a 5% Na₂EDTA (w/v) solution was added to increase the recovery of the macrolide antibiotics.¹⁶

Three different methods were tested. For method I and II, a Strata-X cartridge was used and for method III a Strata-X in tandem with the Strata-X-C cartridge (Strata-X on top of Strata-X-C) was used. For all methods tested, the cartridges were conditioned using 2 × 2.5 mL MeOH and then 2 × 2.5 mL of deionized water containing 50 mM of formic acid and 200 mg L^–1 Na₂EDTA adjusted to pH = 3 with NaOH. Samples were introduced in the cartridges at a flow rate of 2–3 mL min^–1 by negative pressure using Teflon tubes and a vacuum manifold connected to a water trap and a pump. After the cartridges were loaded with the sample, they were dried for 15 min with N₂(g) at maximum pressure (20 in Hg). After drying, the cartridges were eluted in three different ways at a flow rate of 2.5 mL min^–1: method I = 2 × 2.5 mL MeOH : ACN (1 : 1); method II = 2 × 2.5 mL MeOH : ACN : i-PrOH (5 : 5 : 2); and method III = 2 × 2.5 mL MeOH : ACN (1 : 1) for the Strata-X cartridge and 2 × 2.5 mL 5% NH₃–MeOH : ACN (1 : 1) for the Strata-X-C cartridge. The elution solvent was allowed to soak the cartridge for 5 min before each elution.

Eluates were recovered in conical-bottom centrifuge tubes. For method III, the eluates were recovered from both cartridges and were collected in the same conical-bottom centrifuge tube. Eluates were then evaporated to dryness with a gentle stream of N₂(g) at 30 °C and then reconstituted to 250 µL with 0.1% formic acid in 90% H₂O–5% MeOH–5% ACN solution.

For LC, the column temperature was set to 30 °C. Solvent A was 0.1% formic acid–H₂O (pH 2.9) and solvent B was 0.1% formic acid–MeOH : ACN (1 : 1). The injection volume was 20 µL using the full loop mode. Mobile phase flow rate was set to 200 µL min^–1. The following gradient elution program (mobile phase B) was used: 0 min, 10%; 2 min, 10%; 8 min, 20%; 10 min, 20%; 16 min, 60%; 20 min, 95%; 25 min, 95%; 30 min, 10%; 35 min, 10%.

For MS/MS, ionization was performed using electrospray in the positive mode (ESI+). Ion source parameters were optimized for each compound using the Quantum Tune application. ESI optimization solutions were infused with the syringe pump of the TSQ Quantum Ultra and mixed, using a tee, with the LC column flow (200 µL min^–1) before being introduced into the ESI source. Quantum tune was programmed to look for the most intense SRM transitions for each compound. Non-specific SRM transitions (i.e. showing neutral losses of H₂O or CO₂) were not used because many molecules could have similar fragment losses and therefore are not specific to the analyte parent ion. For example, fragmentation of CIP yielded three major product ions: m/z 231, m/z 314 and m/z 288. While m/z 314 was more intense than m/z 288, the former was not used as it represented a loss of water. Only the two more intense and specific SRM transitions were re-optimized to get the best possible source and ion focusing values.

Common source parameters for each compound were averaged to get overall optimal signal intensity and stability and set to the following values: spray voltage, 3500 V; ion transfer capillary temperature, 350 °C; sheath gas pressure, 21 mTorr; auxiliary gas pressure, 4 mTorr; collision gas pressure, 1.5 mTorr and source CID, –12 V. Tube lens and collision energies (CE) are compound-specific and appear in Table 3.

Compound	SMR #1 (m/z)	CE/V	SRM #2 (m/z)	CE/V	Tube lens
PYR	249.10 177.07	40			86
SMX	254.08 92.11	36	254.08 108.10	37	70
DIA	261.15 123.11	34			86
TRI	291.16 123.10	33	291.16 230.17	34	91
CIP	332.16 288.15	27	332.16 231.07	49	82
LOM	352.17 265.13	34			89
LEV	362.17 261.12	35	362.17 221.05	43	92
CLA	748.55 590.36	19	748.55 115.99	35	96
AZI	375.33 82.96	25	749.54 158.04	38	74/112
JOS	828.53 108.87	46	828.53 173.96	47	126

The wastewater interception network of the city of Montréal is divided in two main sectors: north and south. The northern interceptor collects sewage coming from the west and north side of the city while the southern interceptor collects sewage and rainwater from the south side of the city. We collected 24 h composite samples from the WWTP influents (north and south interceptors) and effluent on May 16th 2006. The mean flow for this date was 35.3 m³ s^–1. Matrix characteristics are shown in Table 1. The samples were prepared exactly as above, and they were extracted using method III. To verify the extraction recovery, prior to the extraction 250 µL of the surrogate standard (PYR) was spiked to get a final concentration of 500 ng L^–1. After evaporation, dried residues were reconstituted to 250 µL with 0.1% formic acid in 90% H₂O–5% MeOH–5% ACN solution containing 500 µg L^–1 of the internal standards (DIA, LOM and JOS).

Chromatogram peaks were integrated using the ICIS algorithm of Xcalibur 1.2 by Thermo Fisher Scientific. Signal-to-noise (S/N) ratios were determined with the manual noise region option in Xcalibur. This method was preferred to the default peak to peak method, as it was observed that the latter overestimated S/N ratios. Analyte recoveries were compared using a one-way ANOVA test, and significant differences (p < 0.05) were elucidated with Tukeys b post-hoc test using the statistical software SPSS 13 (Chicago, IL, USA).

Calibration standards used for quantification were prepared by diluting the primary effluent matrices with deionized water by a factor of 10, and then preparing the samples using method III. Five concentration levels were used: 0, 25, 50, 250 and 1000 µg L^–1 and the internal standards and the surrogate were added at a constant concentration of 500 µg L^–1. Calibration curve standards were injected twice during the analysis sequence. The zero level was subtracted from all levels to correct for the anti-infectives already present in the diluted matrix. Internal standards were chosen because of similar structure and physicochemical properties. As these substances are not marketed in Canada, the probability of their presence in the matrix is very low. Their absence in the sample matrices was confirmed by LC/MS/MS analysis of non-fortified samples.

To avoid false blanks, we used a second SRM transition, as well as the area ratio of the two transitions, to confirm the presence of the identified anti-infectives. According to the identification point system proposed by the European Commission,¹⁷ by using one precursor ion and two product ions, each compound earns four identification points (IP), which then fulfils the requirements for identification and confirmation of environmental contaminants.¹⁸ In order to qualify for the IPs required for confirmation, ion ratios must agree within specified tolerances (from ±20% for relative ion intensities >50% to ±50% for relative ion intensities 10%).

Limits of detection (LOD) were determined using the wastewater effluent samples, since field blanks are not available. LODs were estimated by averaging the measured S/N ratio of the analyte peak in the samples (n = 3) and then downscaling to calculate the concentration of the target analyte able to generate a S/N ratio equal to 3. Linear dynamic ranges (LDR) were estimated using calibration curves prepared with the diluted effluent matrix.

The Strata-X cartridge is a reversed-phase sorbent made of a surface modified styrene divinylbenzene polymer. It is designed to enhance retention on polar and aromatic analytes by H-bonding and – interactions. The Strata-X-C sorbent has a benzenesulfonic acid group uniformly bonded on a polymeric surface, and therefore has cation-exchange properties. This type of cartridge has been used previously in trace analysis of anti-infectives¹⁹ but it shows lower recoveries on the less polar compounds. Recovery tests results are shown in Table 4. For method I, analyte recoveries of SMX (50% ± 3), CIP (44% ± 1) and LEV (78.7% ± 0.2) were lower than those of TRI (90% ± 2), AZI (88% ± 6) and CLA (88% ± 6). Elution of the cartridges with a solvent mix containing 16% i-PrOH (method II) provided no significant differences compared to method I.

The tandem SPE approach (method III) significantly improved the recovery of SMX (68% ± 5) TRI (104% ± 4), CIP (76% ± 5) and LEV (97% ± 5) compared to the single cartridge methods (I and II) and was therefore applied to all subsequent analyses. Extraction recoveries of CLA (100% ± 2) and AZI (92% ± 4) with the tandem SPE method were not significantly different to method I and II. Tandem SPE methods have been used to eliminate interferences^20,21 but to the authors knowledge, they have never been applied to more complex water matrices such as raw sewage or primary effluents. The combination of reversed-phase and ion exchange surface chemistry proved to be an effective way to simultaneously extract, from wastewaters, various anti-infectives having different chemical properties such as pK_a and D_ow. The improvement on the recovery of compounds such as the fluoroquinolone antimicrobials CIP and LEV can be explained by their ionic nature. Batt et al.²² reported that the mechanism for the interaction of the fluoroquinolones with the solid phase in the Oasis HLB cartridges (poly[divinylbenzene-co-N-vinylpirrolidone)] is based on electrostatic rather than hydrophobic interactions. At pH = 3, both CIP and LEV are almost completely in their protonated form, which strongly interacts with the ion exchanger benzosulfonic acid group of the Strata-X-C cartridge, therefore increasing overall retention. Interestingly, SMX recovery is also improved using the tandem SPE method. At pH = 3, SMX is mostly in its neutral form (pK_a = 1.7–1.85) and thus its retention can not be improved by ion exchange. However SMX has a benzene ring that can interact via London – interactions with the polymer backbone of the Strata-X-C cartridge, which yields a higher retention. The recovery of the macrolide antibiotics AZI and CLA was very high on all three methods (>85%) and no significant differences were observed among them, which suggests that the main retention mechanism of these compounds is by Van der Waals forces on the Strata-X cartridge.

Tandem mass spectrometry in the SRM mode proved to be highly specific. When two specific SRM transitions are employed, the possibility of false positives is reduced as some matrix interferences co-extracted with the analytes could have the same SRM transition (Fig. 2). Sample SRM transition area ratios were reproducible (RSD < 10%) and differences with SRM transition area ratios of spiked standards in the matrix were not higher than 19%, except for AZI (64%). Instrument response was linear (r² 0.99) in the dynamic range (25–1000 ng L^–1) in spite of the presence of high concentrations of organic as well as inorganic interferences in the matrix. Limits of detection ranged from 0.3 to 22 ng L^–1 (Table 5). Method uncertainty was in most cases <8% RSD, except for CLA and AZI in the south influent.

	Fig. 2 Chromatograms showing two SRM transitions of the studied compounds in treated wastewater. Peaks due to interferences are marked by asterisks (*). The presence of the studied anti-infectives was effectively and unambiguously confirmed by their two specific SRM transitions as well as their area ratio.

Results of the analysis of real samples (Fig. 3) showed that all targeted anti-infectives were found in the Montréal WWTP influents and effluents. The anti-infective found at the highest concentration was CLA in the north influent (263 ± 7 ng L^–1) and LEV was found at the lowest concentration in the treated effluent (39 ± 1 ng L^–1). Concentrations of SMX and TRI in the effluent are similar to previously published data of samples collected at the same WWTP.²³ The reported levels for all anti-infectives are lower or in the same order of magnitude of other Canadian,^24,25 American^20,26 or European cities.^14,18,27,28

	Fig. 3 Occurrence of the targeted anti-infectives in the north influent, south influent and effluent of the Montréal WWTP.

Estimated mean flows of the studied anti-infectives in the St Lawrence River (SMX 340 ± 30, TRI 310 ± 20, CIP 320 ± 10, LEV 118 ± 2, CLA 830 ± 60 and AZI 310 ± 20 g day^–1) show that in spite of the sub-microgram per litre concentrations of anti-infectives found in the Montréal WWTP effluent, large amounts of those products are discharged into the receiving waters of the St Lawrence River. Compared to the mass flow of CIP discharged by Klotten–Opfikon tertiary WWTP in the Glatt River in Switzerland (1.3 ± 0.6 g day^–1), the mean mass flow of CIP coming from the Montréal WWTP is more than 200 times higher. More information is necessary to evaluate the ecoxicological consequences arising from the large amounts of these biologically active compounds being continuously released into the St Lawrence River.

The developed tandem SPE LC/MS/MS method proved to be a rugged and specific method for the extraction and quantification of anti-infectives in wastewaters containing high concentrations of DOC. The tandem SPE method improved anti-infectives recoveries compared to single cartridge methods. Recoveries using tandem SPE cartridges were higher than 75%, except for SMX which was 68%. Detection limits ranged from 0.3 to 22 ng L^–1 and instrument response was linear (r² 0.99) in the dynamic range (25–1000 ng L^–1). The method was successfully applied to real samples from the Montréal wastewater treatment plant. The use of two specific SRM transitions and their area ratios proved to be a reliable and effective way to reduce false positives and confirm the presence of the targeted substances. All the studied anti-infectives were found in the wastewater samples in concentrations ranging from 39 to 276 ng L^–1. Mean flows of anti-infectives were estimated from effluent concentrations and it was found that, despite the low concentrations of these biologically active compounds in the treated sewage, given the significant water flow, large amounts are actually discharged in the receiving waters of the St Lawrence River. More studies are necessary to assay the potential risk and effects of anti-infectives on aquatic biota.

Acknowledgements

We thank the Environment Canada–Conservation Service Research Fund, the National Research and Engineering Research Council of Canada and the Canadian Foundation for Innovation for their financial support. We are also grateful for the help of Luc Tremblay and Christine Yelle at the Montréal WWTP.

References

1		R. Hirsch, T. Ternes, K. Haberer and K. L. Kratz, Sci. Total Environ., 1999, 225, 109–118 [Links].
2		D. W. Kolpin, E. T. Furlong, M. T. Meyer, E. M. Thurman, S. D. Zaugg, L. B. Barber and H. T. Buxton, Environ. Sci. Technol., 2002, 36, 1202–1211 [Links].
3		C. D. Metcalfe, B. G. Koenig, D. T. Bennie, M. Servos, T. A. Ternes and R. Hirsch, Environ. Toxicol. Chem., 2003, 22, 2872–2880 [Links].
4		B. A. Wilson, V. H. Smith, F. Denoyelles and C. K. Larive, Environ. Sci. Technol., 2003, 37, 1713–1719 [Links].
5		S. M. Richards, C. J. Wilson, D. J. Johnson, D. M. Castle, M. Lam, S. A. Mabury, P. K. Sibley and K. R. Solomon, Environ. Toxicol. Chem., 2004, 23, 1035–1042 [Links].
6		C. J. Wilson, R. A. Brain, H. Sanderson, D. J. Johnson, K. T. Bestari, P. K. Sibley and K. R. Solomon, Environ. Sci. Technol., 2004, 38, 6430–6439 [Links].
7		C. D. Watts, B. Crathorne, M. Fielding, C. P. Steel, G. Angeletti and A. Bjorseth, in Analysis of Organic Micropollutants in Water, D. Reidel Publishing Company, Dordrecht, 1983, pp. 120–131.
8		M. L. Richardson and J. M. Bowron, J. Pharm. Pharmacol., 1985, 37, 1–12 [Links].
9		E. M. Golet, A. C. Alder and W. Giger, Environ. Sci. Technol., 2002, 36, 3645–3651 [Links].
10		Top 100 anti-infective medications for Quebec in 2004, IMS Health Canada, Canadian CompuScript, Montréal, QC, 2005.
11		R. Lindberg, P. A. Jarnheimer, B. Olsen, M. Johansson and M. Tysklind, Chemosphere, 2004, 57, 1479–1488 [Links].
12		N. M. Vieno, T. Tuhkanen and L. Kronberg, J. Chromatogr., A, 2006, 1134, 101–111 [Links].
13		E. M. Golet, A. C. Alder, A. Hartmann, T. A. Ternes and W. Giger, Anal. Chem., 2001, 73, 3632–3638 [Links].
14		A. Göbel, C. S. McArdell, M. J. F. Suter and W. Giger, Anal. Chem., 2004, 76, 4756–4764 [Links].
15		P. A. Segura, A. Lajeunesse, P. Turcotte, A. Fürtös, C. Gagnon and S. Sauvé, SETAC St Laurent Regional Chapter 9th Annual Meeting, 2–3 June, Québec, QC, 2005.
16		R. Hirsch, T. A. Ternes, K. Haberer, A. Mehlich, F. Ballwanz and K. L. Kratz, J. Chromatogr., A, 1998, 815, 213–223 [Links].
17		Commission of the European Communities, Official Journal of the European Communities, 2006, L221, 8–36.
18		M. Petrovic and D. Barceló, Anal. Bioanal. Chem., 2006, 385, 422–424 [Links].
19		D. Calamari, E. Zuccato, S. Castiglioni, R. Bagnati and R. Fanelli, Environ. Sci. Technol., 2003, 37, 1241–1248 [Links].
20		J. E. Renew and C. H. Huang, J. Chromatogr., A, 2004, 1042, 113–121 [Links].
21		T. Christian, R. J. Schneider, H. A. Farber, D. Skutlarek, M. T. Meyer and H. E. Goldbach, Acta Hydrochim. Hydrobiol., 2003, 31, 36–44 [Links].
22		A. L. Batt and D. S. Aga, Anal. Chem., 2005, 77, 2940–2947 [Links].
23		F. Gagné, C. Blaise and C. André, Ecotoxicol. Environ. Saf., 2006, 64, 329–336 [Links].
24		C. D. Metcalfe, X. S. Miao, B. G. Koenig and J. Struger, Environ. Toxicol. Chem., 2003, 22, 2881–2889 [Links].
25		X. S. Miao, F. Bishay, M. Chen and C. D. Metcalfe, Environ. Sci. Technol., 2004, 38, 3533–3541 [Links].
26		S. Yang and K. Carlson, Water Res., 2003, 37, 4645–4656 [Links].
27		C. Hartig, T. Storm and M. Jekel, J. Chromatogr., A, 1999, 854, 163–173 [Links].
28		C. S. McArdell, E. Molnar, M. J. F. Suter and W. Giger, Environ. Sci. Technol., 2003, 37, 5479–5486 [Links].
29		H. Sabik, C. Gagnon, F. Houde and C. Deblois, Environ. Forensics, 2004, 5, 61–70 [Links].
30		W. Szczepaniak and A. Szymaski, J. Liq. Chromatogr. Relat. Technol., 2000, 23, 1217–1231 [Links].
31		Z. Qiang and C. Adams, Water Res., 2004, 38, 2874–2890 [Links].
32		T. B. Vree and Y. A. Hekster, Clinical pharmacokinetics of sulfonamides and their metabolities: an encyclopedia, Basel, 1987.
33		Scifinder Scholar 2006, Chemical Abstracts service, American Chemical Society, last accessed 14/02/2007.
34		J. Cao and R. F. Cross, J. Chromatogr., A, 1995, 695, 297–308 [Links].
35		C. A. Kan and M. Petz, J. Agric. Food Chem., 2000, 48, 6397–6403 [Links].
36		P. K. Martindale, The Complete Drug Reference, Pharmaceutical Press, London, 1999.
37		A. I. Drakopoulos and P. C. Ioannou, Anal. Chim. Acta, 1997, 354, 197–204 [Links].
38		D. L. Ross and C. M. Riley, Int. J. Pharm., 1992, 83, 267–272 [Links].
39		H. R. Park, T. H. Kim and K. M. Bark, Eur. J. Med. Chem., 2002, 37, 443–460 [Links].
40		A. J. Bryskier, J. P. Butzler, H. C. Neu and P. M. Tulkens, Macrolides: Chemistry, pharmacology and clinical uses, Arnette Blackwell, Paris, 1993.
41		J. W. McFarland, C. M. Berger, S. A. Froshauer, S. F. Hayashi, S. J. Hecker, B. H. Jaynes, M. R. Jefson, B. J. Kamicker, C. A. Lipinski, K. M. Lundy, C. P. Reese and C. B. Vu, J. Med. Chem., 1997, 40, 1340–1346 [Links].