Spatiotemporal distribution of pharmaceuticals in the Douro River estuary (Portugal)

Spatiotemporal distribution of pharmaceuticals in the Douro River estuary (Portugal)

Science of the Total Environment 408 (2010) 5513–5520 Contents lists available at ScienceDirect Science of the Total Environment j o u r n a l h o m...

712KB Sizes 0 Downloads 20 Views

Science of the Total Environment 408 (2010) 5513–5520

Contents lists available at ScienceDirect

Science of the Total Environment j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c i t o t e n v

Spatiotemporal distribution of pharmaceuticals in the Douro River estuary (Portugal) Tânia Vieira Madureira a,b,c,d,e, Juliana Cristina Barreiro e, Maria João Rocha a,d, Eduardo Rocha b,d, Quezia Bezerra Cass e, Maria Elizabeth Tiritan a,c,⁎ a

Health Sciences Research Center of the Superior Institute Health Sciences North, Gandra, Paredes, Portugal Institute of Biomedical Sciences Abel Salazar (ICBAS), University of Porto, Portugal Medicinal Chemistry Centre (CEQUIMED-UP), University of Porto, Portugal d Interdisciplinary Centre of Marine and Environmental Research (CIIMAR), CIMAR Associated Laboratory, Porto, Portugal e Chemistry Department, Federal University of São Carlos, P.O. BOX 676, São Carlos, 13565-905, SP, Brazil b c

a r t i c l e

i n f o

Article history: Received 23 October 2009 Received in revised form 2 July 2010 Accepted 28 July 2010 Available online 21 August 2010 Keywords: Toxicological risk Pharmaceuticals Douro River estuary Surface water Environmental monitoring

a b s t r a c t The amount and distribution of six pharmaceutical compounds belonging to distinct therapeutic classes were investigated along the navigation channel of the Douro River estuary. Distinct spatial and temporal trends were considered and a total of 87 water samples were pre-concentrated by solid-phase extraction (SPE) and analyzed by liquid chromatography tandem mass spectrometry (LC-MS/MS) with an ion trap (IT) analyzer and electrospray ionization (ESI). The maximum concentrations found were 178 ng/L for carbamazepine, 3.65 ng/L for diazepam, 70.3 ng/L for fenofibric acid, 3.18 ng/L for propranolol, 15.7 ng/L for trimethoprim and 53.3 ng/L for sulfamethoxazole. Carbamazepine was the most ubiquitous compound with 100% positive detection frequency followed by propranolol (38%), trimethoprim (34%) and sulfamethoxazole (33%). The pharmaceutical compounds were quantified at higher levels in the lower stretch of the estuary, especially near the wastewater treatment plant (WWTP). The data proves that pollution of the Douro River estuary by pharmaceuticals is consistent and is occurring in a fairly constant manner in time, covering a wide area and displaying hot-spots. Individually, the concentration levels are not likely to cause acute effects, based on reference experimental data. However, the fact that complex mixtures exist gives cause for concern as regards potentially relevant toxicological risks. The study points out the need for continuous monitoring of contamination levels not only in the Douro River estuary but also in other major estuaries. Finally, the scenario supports the need for experimental studies on toxicological impacts on aquatic organisms at environmentally relevant concentrations. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Pharmaceutical compounds have been recognized as priority concern micropollutants in the contamination of distinct aquatic compartments (Khetan and Collins, 2007; Kümmerer, 2009). Complex mixtures of these compounds along with their metabolites are continually introduced into the environment, via distinct pathways, such as the disposal of unused, expired medicines or by untreated discharges and/or wastewater treatment plant (WWTP) effluents, after excretion by humans (Bound and Voulvoulis, 2005). Effluents from hospitals and drug manufacture (Langford and Thomas, 2009; Larsson et al., 2007), land application of biosolids, commonly used in agriculture as fertilizers, and water reuse (irrigation) (Topp et al., 2008; Xia et al., 2005) are other important pathways that contribute to the continuous introduction of pharmaceuticals into aquatic ⁎ Corresponding author. Health Sciences Research Center of the Superior Institute Health Sciences North, Gandra, Paredes, Portugal. E-mail addresses: [email protected] (T.V. Madureira), [email protected] (M.E. Tiritan). 0048-9697/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2010.07.069

environments. Additional inputs can also derive from therapeutic use in animal production, such as in aquaculture (Cole et al., 2009). Since WWTPs were not designed to degrade organic molecules at low concentration levels, the introduction of many human pharmaceuticals into aquatic environments is commonly observed (Williams, 2003). WWTP elimination rates are often incomplete and can vary greatly when considering distinct pharmaceuticals and/or different WWTPs (Gros et al., 2010; Santos et al., 2009). Thus, the elimination rates of some pharmaceuticals are only about 7%, such as for carbamazepine, while others reach 96% efficiency (Ternes, 1998). Moreover, pharmaceuticals have a wide persistence range and inherent bioactive properties which explains their relevance in terms of environmental contamination (Fent et al., 2006; Williams, 2003). They may induce a variety of physiological changes, reversible or not, in non-target aquatic organisms (Daughton and Ternes, 1999; Khetan and Collins, 2007; Morley, 2009; Santos et al., 2010). Ecotoxicological studies of distinct aquatic species involving different therapeutic classes of pharmaceuticals such as antidepressants, betablockers and lipid regulators have been under investigation (Brooks et al., 2003; Huggett et al., 2002; Isidori et al., 2007). Thus, considering


T.V. Madureira et al. / Science of the Total Environment 408 (2010) 5513–5520

the impact on aquatic biota as well as on human health, it is extremely important to carry out temporal and/or spatial field studies to evaluate the fate and dynamic distribution of pharmaceuticals in aquatic environments. This work presents the first study concerning the spatiotemporal distribution of six emerging pharmaceutical compounds along the Douro River estuary, Portugal. The Douro River is one of the longest rivers of the Iberian Peninsula (930 km), and it has a 98,000 km2 watershed shared between Spain and Portugal. In Portugal, the river's surface and underground water resources are used for the production of hydroelectricity (53% of the national total), irrigation, industrial purposes and to supply drinking water to half of the population residing in the metropolitan area of Porto (Cortes, 2009). The first 9 km of the mouth of the estuary is densely populated since two major cities, Porto and Gaia, are located on the north and south banks, respectively. Overall, anthropogenic stress on the Douro River estuary results from the industrial development and also from the WWTP effluents that are discharged into the estuary or its tributaries. A recent study has demonstrated sewage dissolved organic carbon input in the lower and middle stretches of the estuary (Magalhães et al., 2008). The presence throughout the estuary of endocrine disrupting compounds (Ribeiro et al., 2009), heavy metals (Mucha et al., 2005) and other organic pollutants such as polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs) and organochlorine pesticides (OCPs) has already been reported in this area (Ferreira et al., 2004, 2006). On the other hand, to our knowledge, the systematic quantification of pharmaceuticals, considering spatial and temporal distributions, has never been carried out in spite of the evidence of contamination risk. The pharmaceuticals selected included trimethoprim (TMP), propranolol (PHO), sulfamethoxazole (SMX), carbamazepine (CBZ), diazepam (DZ) and the active metabolite of fenofibrate (F), fenofibric acid (FA). The selected compounds belong to various therapeutic classes with distinct physico-chemical properties, environmental behaviour and persistence, in order to have a heterogeneous mixture of pharmaceuticals which have been commonly reported by other studies (Mompelat et al., 2009; Santos et al., 2010). In addition, the consumption of these pharmaceuticals in Portugal during 2001 and 2005 was also taken into consideration (INFARMED, 2009). Some of these compounds were selected to be monitored in an EU-wide survey of polar organic persistent pollutants in several European rivers (Loos et al., 2009). Moreover, a recent microbial breakthrough study has also considered three of these compounds (CBZ, SMX and

TMP) as very refractory and, as such, excellent candidates for the identification of wastewater contamination (Benotti and Brownawell, 2009). Accordingly, the main aims of this work were to: (1) identify and quantify the target pharmaceuticals in surface waters of the Douro River estuary; (2) determine the distribution of the compounds throughout the estuary as a function of time and space to assess potential pollution sources; (3) contribute to a broad understanding of the estuarine water quality in order to better predict the extent of aquatic ecosystem risks relative to pharmaceutical pollution; and (4) obtain data for establishing toxicological studies on the potentially adverse biological effects of environmentally relevant concentrations.

2. Materials and methods 2.1. Chemicals and materials Trimethoprim (TMP), propranolol hydrochloride (PHO), sulfamethoxazole (SMX), carbamazepine (CBZ), diazepam (DZ) and fenofibrate (F) were purchased from Sigma-Aldrich (Steinheim, Germany). Fenofibric acid (FA) was prepared by hydrolysis of fenofibrate as described elsewhere (El-Gindy et al., 2005; Rath et al., 2005). Internal standards (IS), [13C1,15N1]-carbamazepine and [d4]sulfamethoxazole were obtained from Sigma-Aldrich (Steinheim, Germany) and Toronto Research Chemicals Inc. (North York, Canada), respectively. All pharmaceutical standards were of analytical grade with purity greater than 98%. Individual standard solutions at 1 mg/ mL were prepared by weighing and dissolving 1 mg of standard in ethanol and stored at −20 °C in the dark. For the IS, solutions at 10 μg/ mL were prepared in ethanol. Standard mixtures, at different concentrations, used during the validation procedure were prepared by appropriate dilution of the individual stock solutions as described in a previous work (Madureira et al., 2009). Analytical grade formic acid (98%) was obtained from Fluka (Steinheim, Germany) and ultrapure water was purified through a Milli-Q system (Millipore, São Paulo, Brazil). All other solvents used were HPLC grade and supplied by J.T. Baker (Philipsburg, USA, PA). The Oasis® HLB (Hydrophilic–Lipophilic Balance) cartridges, 500 mg, 12 cc from Waters Corporation (Milford, MA, USA) were used for solid-phase extraction (SPE). The 0.45 μm glass fibre filters were purchased from Millipore (Ireland).

Fig. 1. Location of the sampling sites within the Douro River estuary, Portugal.

T.V. Madureira et al. / Science of the Total Environment 408 (2010) 5513–5520

2.2. Study area The Douro River drains into the Atlantic Ocean at 41°08′N and 8°40′W and its estuary is controlled upstream by the Crestuma–Lever dam at about 21.6 km from the mouth. The Douro River estuary has an average depth of 8 m and the tides are semidiurnal ranging from 2 to 4 m. The water level in the whole estuary varies almost simultaneously during flood and ebb tides with only 1 h difference between the mouth and the dam. The residence time changes between 0.3 and 16.5 days with the increase or decrease of freshwater flow, respectively. The freshwater flow controlled by the Crestuma–Lever dam mainly determines the water circulation and it ranges from 0 m3 s−1 in summer to 13,000 m3 s−1 in winter flood events with an average discharge of 505 m3 s−1 (Bordalo and Vieira, 2005). However, a large inter-annual variability in the flow can occur depending on whether it is a wet or dry year (Azevedo et al., 2006). The Douro River estuary receives, directly or indirectly, effluents from eight WWTPs (Fig. 1). Two of the largest WWTPs (Sobreiras and Freixo) are located in the lower and medium stretch of the estuary. The Sobreiras WWTP treats domestic wastewater from the western part of Porto city and was sized to serve an equivalent population of 200,000 inhabitants. The average daily flow corresponds to 54,000 m3. The Freixo WWTP serves a population of 170,000 inhabitants and has an average daily flow of 35,900 m3.


filtered through 0.45 μm glass fibre filters to remove suspended particles. Each filter was washed with approximately 5 mL of methanol and this volume was added to the filtrate. First, all samples were adjusted to pH 7 with H2SO4 (2 mol/L) and then maintained at ±4 °C until extracted within a maximum of 72 h after collection. 2.4. Sample pre-concentration All water samples were extracted according to a SPE method previously established by our group (Madureira et al., 2010). Briefly, SPE cartridges were conditioned with 32 mL of dichloromethane, 32 mL of methanol and then equilibrated with 32 mL of ultrapure water, at a flow rate of 1 mL/min. Subsequently, surface water samples (2 L) were percolated through the cartridges at a flow rate of 10 mL/min. The cartridges were then rinsed with 32 mL of ultrapure water and vacuum dried for approximately 30 min. After drying, SPE cartridges were wrapped in aluminium foil and stored at 4 °C until eluted. The elution was performed with 32 mL methanol/dichloromethane (70:30, v/v) at 1 mL/min into a test tube containing 4 μL of each internal standard (10 μg/mL). Eluates were evaporated to dryness in a thermostatic bath at 40 °C under a gentle nitrogen stream and redissolved with 400 μL of ethanol obtaining 5000-fold pre-concentration. 2.5. Sample stability

2.3. Sample collection and site description Water samples (2 L) were collected at eleven sampling stations in the Douro River estuary, along the navigation channel of the waterway, from the mouth to the Crestuma–Lever dam, as shown in Fig. 1. All samples were collected in the middle of the estuary with the exception of samples from sites 2, 4, 8 and 10 that were sampled closer to the banks of the estuary. Sampling was performed from October 2007 to July 2008, in four different periods, representing typical seasonal conditions, during ebb and flood tides. 22 samples were collected during each seasonal survey, with the exception of autumn (the flood tide sample from site 2 was not collected), yielding a total of 87 estuarine water samples. Several duplicate samples were also collected at specific sampling sites (2, 8 and 10) in order to perform a comparative study of refrigerated and frozen pre-treated samples. Water samples were collected in 2.5 L amber glass bottles, which were rinsed in the laboratory with ultrapure water, and subsequently with the water sample at the sampling site. Surface waters were sampled from a depth of approximately 1 m using a peristaltic pump (Global Water, model: WS 3000, California, USA). The pH, conductivity, salinity, dissolved oxygen and water temperature were measured in situ using a portable Consort apparatus (Table 1). Upon collection, the samples were kept refrigerated (±4 °C), transported in the dark to the laboratory and vacuum

The stability of pharmaceutical compounds stored in the cartridges was evaluated using spiked water samples (80 ng/mL — TMP, 8 ng/mL — PHO, 70 ng/mL — SMX, 6 ng/mL — CBZ, 16 ng/mL — DZ and 32 ng/mL — FA) prepared in duplicate at a concentration unknown to the analyst and stored in the SPE cartridges over a period of six months. Another stability experiment was carried out using water samples collected on the same day and at the same sampling site to test different storage conditions (4 °C and −20 °C). In order to have a reliable comparison between samples, they were extracted simultaneously. All compounds were stable during the storage period in the cartridges since the accuracy obtained between stored and nonstored samples ranged from 90.4 to 119% with RSD values lower than 8.45%. Considering all the water samples stored at 4 °C and −20 °C, the RSD values obtained ranged from 2.11 to 22.4%. 2.6. Liquid chromatography tandem mass spectrometry (LC-MS/MS) The HPLC analyses were performed in accordance with a previously validated method (Madureira et al., 2009). A Shimadzu (Kyoto, Japan) system equipped with two LC-20AD pumps, SIL-20A autosampler, DGU-20A5 degasser, UV/Vis SPD-20A detector and a CBM-20A interface was used. The pharmaceutical compounds were separated on a Shimadzu C18 endcapped column (150 × 2.1 mm,

Table 1 Range (min–max) obtained for the physico-chemical parameters of water samples from the Douro River estuary, Portugal. Sampling stations

Physico-chemical parameters Temperature (°C)

Conductivity (mS/cm)

Salinity (‰)*


Oxygen saturation (%)

Concentration of dissolved oxygen (mg/L)

Site Site Site Site Site Site Site Site Site Site Site

10.0–27.1 10.0–28.4 9.0–25.4 10.0–26.0 9.0–26.4 9.0–27.0 10.0–26.0 9.0–26.0 9.0–26.0 10.0–26.0 10.0–26.0

13.1–43.5 13.6–29.5 7.5–22.7 3.7–19.0 3.1–12.0 1.2–10.4 0.4–3.9 0.2–1.4 0.2–0.8 0.2–1.4 0.2–1.1

8.1–31.2 7.9–20.1 4.2–10.1 1.8–8.7 0.7–5.3 0.6–4.8 0–2.6 0–0.7 0–0.4 0–0.7 0–0.5

7.1–8.5 6.9–8.3 6.9–8.2 7.1–8.2 7.1–8.1 7.2–8.1 7.2–8.1 7.3–8.1 7.4–8.6 7.4–8.1 7.6–8.2

31.9–70.0 31.0–71.0 30.8–73.1 32.2–88.0 31.7–83.3 31.4–70.1 31.8–71.1 32.7–74.0 30.4–79.2 30.6–75.9 30.3–72.0

3.0–6.4 2.7–6.7 2.8–6.8 2.9–8.4 2.8–7.8 3.0–6.7 2.9–6.8 2.8–6.8 3.1–7.3 3.0–7.2 2.9–7.2

1 2 3 4 5 6 7 8 9 10 11

⁎ Salinity was considered 0‰ when the value was below 0.2‰.


T.V. Madureira et al. / Science of the Total Environment 408 (2010) 5513–5520

5 μm) with a 0.1% aqueous solution of formic acid, v/v (solvent A) and acetonitrile with 0.1% formic acid, v/v (solvent B) as mobile phases. The gradient elution was linear from 10 to 65% of B in 20 min and then from 65 to 85% of B in 4 min with a flow rate of 0.2 mL/min. The equilibration time was 5 min and the injection volume was set to 30 μL. The LC flow was split and only 100 μL/min was directed into the mass spectrometer. For mass spectrometry analysis, an Esquire 6000 ion trap (IT) mass spectrometer (Bruker Daltonics, Germany) with the electrospray ionization (ESI) source operating in positive ion mode was used. The protonated molecular ion was selected as precursor ion for all compounds owing to its stability and abundance. Detection was performed in multiple reaction monitoring (MRM) mode and two MRM transitions were monitored (Table 2). The first transition was used for quantification and the second one for confirmation purposes. The quantification procedure was described in detail in a previous work (Madureira et al., 2009). Briefly, SMX and CBZ were quantified using isotopic internal standards while FA, PHO, DZ and TMP were quantified by external calibration. The confirmation of the compounds' identity in water samples from the Douro River estuary was performed in compliance with European regulations (Commission Decision 2002/657/EC). Thus, the LC retention times (tR) and the area ratio of the two product ions in each estuarine water sample were compared with those of standard solutions analyzed under identical conditions. All water samples were injected in duplicate. The method detection (MDL) and quantification limits (MQL) were below 1.30 and 3.24 ng/L, respectively, with the exception of SMX (4.40 and 6.60 ng/L, respectively). The correlation coefficients (r) of the calibration curves were always higher than 0.99. The characteristics of the analytical method are summarized in Table 2.

3. Results and discussion 3.1. Pharmaceutical levels in surface waters of the Douro River estuary All of the pharmaceutical compounds monitored were found at least once in the Douro River estuary. CBZ was quantified in 100% of estuarine water samples with concentrations ranging from 0.37 ng/L to 178 ng/L (Fig. 2). These results are in accordance with the data reported by different researchers for surface water from the Otonabee River, Tennessee River, Somes River and Aa and Aabach tributaries of Lake Greifen in which CBZ was detected at 0.7 ng/L, 2.9–23.1 ng/L, 65– 75 ng/L and 30–250 ng/L, respectively (Conley et al., 2008; Miao and Metcalfe, 2003; Moldovan, 2006; Öllers et al., 2001). Concentrations of up to 1 μg/L (Heberer, 2002) and even 6.72 μg/L (Feitosa-Felizzola and Chiron, 2009) of CBZ have already been detected in surface water samples from Berlin and from a small Mediterranean watercourse, Arc River. CBZ accounted for 26% of the anticonvulsants consumed in Portugal from 2001 to 2005 (INFARMED, 2009). Furthermore, its low removal efficiency by WWTPs (≤10%) is well reported (Ternes, 1998), and there are also reports of a slight increase when comparing CBZ concentration levels of WWTP influents and effluents, as a result of cleavage of the glucuronide conjugate (Vieno et al., 2006). These facts along with its resistance to photodegradation in surface waters, when compared to other compounds (Andreozzi et al., 2003; Yamamoto et al., 2009), explain the widespread presence of CBZ in the Douro River

2.7. Statistical analyses The frequency of positive detection in percentage (%), the mean and standard deviation measures of replicate concentrations were calculated with Excel software (Microsoft). To confirm or disprove the visual/graphical data patterns, comparisons between situations were studied by ANOVA with post-hoc testing among pairs being done with the Newman–Keuls test. Whenever homoscedasticity did not exist (as judged by Levene's test) or could not be generated by data transformation, a non-parametric approach using Kruskal–Wallis ANOVA by ranks followed by post-hoc multiple comparisons (which gave virtually all the same differences as obtained from the parametric approach) was used. Data was analyzed both by grouping all sites or seasons, or by sites grouped by salinity influence, or even site by site whenever there was sufficient data per site (grouped by seasons). For a single compound, there was not enough data to enable statistical comparisons of ebb and flood values in a single site within a particular season. All tests were made with the STATISTICA 8.0 software (StatSoft). The p-level was set at 0.05.

Table 2 Summary of the method performance parameters: range (ng/mL), retention time (tR), multiple reaction monitoring transitions (MRM), method detection limit (MDL) and method quantification limit (MQL). Compound

Range (ng/mL)

tR MRM transitions MDL MQL (min) (m/z) (ng/L) (ng/L)

Trimethoprim Propranolol Sulfamethoxazole [d4]-sulfamethoxazole (IS) Carbamazepine [13C1, 15N1]-carbamazepine (IS) Diazepam Fenofibric acid

16.2–230 6.7 0.60–8.10 11.3 33.0–446 12.3 12.3 0.54–7.30 15.2 15.2 13.0–176 19.0 6.00–81.0 22.8

291 N 230; 260 N 183; 254 N 156; 258 N 160; 237 N 194; 239 N 194; 285 N 257; 319 N 233;

123 116 108 112 192 192 154 139

1.25 0.03 4.40

3.24 0.12 6.60



1.30 0.20

2.60 1.20

Fig. 2. Concentrations of CBZ (ng/L) along the distinct sampling sites of the Douro River estuary during the four seasons in ebb (a) and flood (b) tides.

T.V. Madureira et al. / Science of the Total Environment 408 (2010) 5513–5520

estuarine areas. Some authors have already reported the persistent pattern of CBZ in Jamaica Bay (Benotti and Brownawell, 2007) and Tamagawa estuary (Nakada et al., 2008) suggesting the usefulness of this compound as a conservative water soluble molecular marker of wastewater contamination (Nakada et al., 2008). SMX and TMP are commercially available as antibacterial compounds generally used in association with each other (5:1 ratio). Both compounds were found with similar detection frequencies in the sample sites evaluated (33% — SMX and 34% — TMP) but the concentration range for TMP varied from 3.89 ng/L to 15.7 ng/L, while the SMX concentrations were higher (9.14 ng/L to 53.3 ng/L). Although there is a correlation between the concentrations of these compounds and the ratio in which they are available in pharmaceutical formulations, no direct conclusions should be drawn as multiplefactors can influence the concentrations found in aquatic environments. For example, approximately 15% of SMX and 60% of TMP are excreted in unchanged form by humans (Hirsch et al., 1999). The removal efficiencies of TMP (60%) and SMX (20%) in WWTPs also differ as reported in the literature (Brown et al., 2006). The concentrations found for these pharmaceuticals are in agreement with published data for surface waters (Kasprzyk-Hordern et al., 2007). Some publications reported higher concentrations for SMX in surface waters of the Rio Grande in New Mexico (300 ng/L) (Brown et al., 2006) and Llobregat River (1488 ng/L) (Díaz-Cruz et al., 2008). However, in this study such high concentrations were not noted, possibly due to a lower SMX consumption rate in human and veterinary therapies in Portugal and/or other causes, such as comparatively better sewage treatment. Additionally, a higher dilution rate in the Douro River estuary due to a diversity of factors such as tidal influence, tributary flow and dam discharge can also explain the results obtained. PHO was detected in 38% of the water samples at a maximum concentration of 3.18 ng/L. Research in the Tyne River (35–107 ng/L) (Roberts and Thomas, 2006) found higher concentrations than those reported in this study, but the data obtained for PHO in the Ouse River (3.9–28.5 ng/L) (Zhang and Zhou, 2007) is closer to the Douro River situation. The values generally found for PHO are low due to its low excretion rate in an unchanged form (b1%) (Ternes, 1998), the high elimination rate through WWTP treatment (96%) (Ternes, 1998) and to the easy photodegradation (half-life b 24 h) (Yamamoto et al., 2009). Although this beta-blocker is commonly used in Portugal, the low levels found in the Douro River estuary may be explained by a high dilution rate in this study area in accordance with the pattern obtained for SMX. The concentrations of FA in the Douro River estuary were between 1.48 ng/L and 70.3 ng/L (32% of detection frequency). The results are in line with the recent values found in the Rhine River (Sacher et al., 2008). DZ, an anxiolytic drug, had a detection frequency in the Douro River estuary samples of just 13% and it was found only once above the MQL (3.65 ng/L), which contrasts with the higher concentrations reported in the Somes River (~ 30 ng/L) (Moldovan, 2006).


depending on proximity to the river mouth and whether the measurement was made at ebb or flood tide. Conductivity followed the same pattern of variation throughout the Douro River estuary with higher salinity and conductivity values at sites 1 and 2 during flood tides. In some cases, at the sampling sites located 20 km from the Atlantic Ocean (sites 9, 10 and 11), the salinity did not reach measurable values. Concentrations of dissolved oxygen ranged from 2.7 to 8.4 mg/L corresponding to an oxygen saturation that varied from 30.3 to 88.0%, with no significant variations between the different sampling sites during each seasonal survey. Considering the detection and the concentrations found for the six compounds at the eleven sampling sites in the Douro River estuary, the locations most affected by pharmaceutical contamination can be identified. The lower stretch of the Douro River estuary (sites 1, 2 and 3) corresponds to an urbanized area that is consequently more vulnerable to human influence. In fact, at site 2 all compounds were quantified at the highest concentrations, demonstrating that these compounds mainly derived from human use and are influenced by the location of a WWTP (Figs. 1 and 3). The concentrations measured for the majority of pharmaceuticals were higher immediately downstream of the WWTP (site 2, Fig. 1) when compared to an upstream location (site 3, Fig. 1), as shown by the example of CBZ (Fig. 2). This result agrees with a series of previously published studies that have also reported a correlation between the spatial location of discharges from municipal WWTPs and compound concentrations (Heberer, 2002; Tamtam et al., 2008). WWTPs as a source of pharmaceuticals release into the aquatic environments are well recognized, due to the fact that conventional treatments are not specifically designed to degrade and/or remove pharmaceuticals.

3.2. Inter-site and seasonal differences The sampling sites were selected in order to obtain a representative number of samples that would characterize the estuarine area in terms of distinct contamination sources, such as urban, industrial, agricultural or WWTP effluents. The quantitative results obtained for the distinct pharmaceutical compounds along with the information provided by the physico-chemical parameters demonstrated that the different sampling sites selected in the Douro River estuary presented high variability in terms of characteristics and water quality (Table 1). The water temperature ranged from 9.0 to 28.4 °C in winter and early autumn, respectively, and the pH values were between 6.9 and 8.6 among the various samples and sampling sites. Concerning the salinity of the estuarine water, distinct values were measured

Fig. 3. Pharmaceutical concentrations at site 2 during the monitoring program in ebb (a) and flood (b) tides.


T.V. Madureira et al. / Science of the Total Environment 408 (2010) 5513–5520

Beyond site 2, the concentrations were higher at sites 4, 8 and 9 when compared to the other sampling sites, as can be noted by the example of CBZ (Fig. 2). Although there is no WWTP or tributary discharge close to site 4, the CBZ concentrations were higher than at site 5 since an untreated discharge is located in the north bank of the estuary close to site 4. Site 8 also showed high concentrations of CBZ, which may be due to the influence of the Sousa River, a well known polluted tributary of the Douro River estuary. As sites 2, 4 and 8 are located close to the banks of the estuary this could explain the higher concentrations found in comparison with the other sampling sites that were distributed in the middle of the estuary. The higher concentrations found at site 9 may arise from the influence of the WWTP located upstream; however, such values were not found at site 10. For the evaluation of the results pertaining to the distribution of the pharmaceuticals along the Douro River estuary, the sampling sites were grouped by salinity influence as follows: site 1 to site 3, site 4 to site 6 and site 7 to site 11. Although there is a tendency for high concentrations in the lower estuarine stretch (sites 1 to 3) for all compounds, in fact significant differences can only be proved for CBZ, PHO and SMX as shown in Fig. 4. Regarding all sampling sites in the four seasons and at the two tides, evaluation of the lower stretch of the Douro River estuary showed the following mean concentrations: CBZ — 15.4 ng/L, FA — 16.8 ng/L, PHO — 1.33 ng/L, SMX — 30.5 ng/L, TMP — 11.5 ng/L and DZ was found only in one sample at 3.65 ng/L. The upper stretch of the estuary is also influenced by the presence of two WWTPs but this area is not as densely populated as the lower stretch and, consequently, there is a smaller volume of wastewater production/discharge. Statistically, no relevant differences were found between the two flow conditions (ebb and flood tides) and between the four seasonal surveys carried out. In fact when considering only the consumption rate of the pharmaceuticals, seasonal variations were not expected to occur, since these compounds are administered continuously (PHO, FA, and CBZ) or at most occasionally (TMP and SMX), but with no specific association with the season of the year. However, other factors such as biodegradation, photodegradation, sediment adsorption, adsorption to suspended solids and water flow variations can also influence the differences in the concentrations found. With regard to site 2, higher concentrations were found for CBZ during dry seasons (summer and autumn) compared with the wet seasons (winter and spring). For TMP, SMX and FA this behaviour was not observed and the higher concentrations were generally found during winter with the exception of SMX in ebb tide (Fig. 3). The slow rate of photo and microbial degradation (Andreozzi et al., 2003; Benotti and Brownawell, 2009) along with the negligible sorption of CBZ onto sludge (Ternes et al., 2004) indicates that this compound once released into the aquatic environment will undergo low elimination rates. So, the high concentrations found for CBZ during the dry seasons may be related mainly with a decrease in the flow of the estuary. In the case of TMP and SMX, although they are hydrophilic enough to be transferred into the aquatic environment with negligible sorption to the sludge biomass and present a great potential to resist biodegradation in WWTPs (Li and Zhang, 2010) and/or microbial degradation in estuarine and coastal surface waters (Benotti and Brownawell, 2009), they can also be subject to phototransformation (Sirtori et al., 2010; Trovó et al., 2009). With regard to FA, fast and complete degradation by sunlight has been reported (Cermola et al., 2005). Thus, for TMP, SMX and FA a possible explanation for the low concentrations obtained during the dry seasons may be a high elimination rate by photodegradation, which in these cases, compensates the eventual decrease in the estuarine flow in autumn and summer. Additionally, the temperatures during the winter are low and, consequently, these compounds were subject to decreased biodegradation by the WWTPs and photodegradation. For PHO only a slight increase was registered in the summer at ebb tide,

Fig. 4. Box–whisker plot (made in the variant of mean + SE + 1.96 SE) grouping sampling sites: 1–3; 4–6; 7–11 for a) CBZ, b) SMX and c) PHO.

however, at flood tide this was not observed (Fig. 3). This compound has high photolability (Yamamoto et al., 2009) and, consequently, it was expected to be more easily degraded during the hotter and drier seasons. The tidal influence on the pharmaceutical occurrences and distribution along the Douro River estuary was also evaluated in order to verify the possibility of a dilution effect on the concentrations along the sampling sites. The sampling sites from 7 to 11 are not

T.V. Madureira et al. / Science of the Total Environment 408 (2010) 5513–5520

influenced by sea water given the salinity values obtained (Table 1); so the concentrations obtained in ebb and flood tides should be the same. However, the concentrations found varied in these sites (Fig. 2), which leads to the conclusion that other factors such as the discharge of freshwater, the variability of location and timing inputs in the Douro River estuary are also interfering. Overall, seasonal and tidal variations did not follow a clear pattern for all compounds studied due to the multitude of parameters occurring in time and space, which complicates the interpretation of the results and makes it difficult to characterize this estuary. The variability of data impairs significance and this can only be revealed with longer systematic surveys involving a higher number of replicate samples. 4. Conclusion All pharmaceutical compounds were quantified in the Douro River estuary at least once at ng/L levels. However, CBZ was the most ubiquitous compound and the one quantified at the highest concentration and at all sampling sites, confirming the persistence of this compound as already indicated elsewhere (Daneshvar et al., 2010). Although the highest concentration for CBZ (178 ng/L) found in the Douro River estuary is lower than the lowest observed effect concentration (LOEC) described in a study (1 μg/L) using rainbow trout (Oncorhynchus mykiss) and common carp (Cyprinus carpio) (Triebskorn et al., 2007), the safety margin may be narrowed if a mixture of pharmaceuticals is considered. In fact, the results obtained in the Douro River estuary indicate that other pharmaceutical compounds not analyzed in this study are present in the study area, and therefore, there is a potential risk of individual and/or mixtures of pharmaceuticals interfering with the homeostasis of local aquatic species. The spatial distribution of the pharmaceuticals along the Douro River estuary showed an overall trend of higher concentrations in the lower stretch, which corresponds to the most urbanized area. This work reiterates that the “hot-spot” sites are strongly influenced by the location of WWTPs, direct discharge of illegal untreated effluents and to the influence of contaminated water from the tributaries. The data obtained with this monitoring study in the Douro River estuary contributes to the information available on the degree of pharmaceutical contamination, distribution and location of the problematic sources and also represents an important baseline for future comprehensive investigations in this complex watershed. Acknowledgments The work has been supported by Fundação para a Ciência e Tecnologia — FCT, Portugal (PhD grant SFRH/BD/31382/2006 attributed to Tânia Vieira Madureira), Conselho Nacional de Desenvolvimento Científico e Tecnológico — CNPq, Brazil and Cooperativa de Ensino Superior, Politécnico e Universitário, CRL (CESPU)/ISCS-N. The authors also wish to acknowledge the support from Capitania do Porto do Douro in the collection of the Douro River water samples and INFARMED for the information given about the consumption rates of the pharmaceuticals in Portugal. References Andreozzi R, Raffaele M, Nicklas P. Pharmaceuticals in STP effluents and their solar photodegradation in aquatic environment. Chemosphere 2003;50:1319–30. Azevedo IC, Duarte PM, Bordalo AA. Pelagic metabolism of the Douro estuary (Portugal): factors controlling primary production. Estuar Coast Shelf Sci 2006;69:133–46. Benotti MJ, Brownawell BJ. Distributions of pharmaceuticals in an urban estuary during both dry and wet-weather conditions. Environ Sci Technol 2007;41:5795–802. Benotti MJ, Brownawell BJ. Microbial degradation of pharmaceuticals in estuarine and coastal seawater. Environ Pollut 2009;157:994-1002.


Bordalo AA, Vieira MEC. Spatial variability of phytoplankton, bacteria and viruses in the mesotidal salt wedge Douro Estuary (Portugal). Estuar Coast Shelf Sci 2005;63: 143–54. Bound JP, Voulvoulis N. Household disposal of pharmaceuticals as a pathway for aquatic contamination in the United Kingdom. Environ Health Perspect 2005;113: 1705–11. Brooks BW, Foran CM, Richards SM, Weston J, Turner PK, Stanley JK, et al. Aquatic ecotoxicology of fluoxetine. Toxicol Lett 2003;142:169–83. Brown KD, Kulis J, Thomson B, Chapman TH, Mawhinney DB. Occurrence of antibiotics in hospital, residential, and dairy effluent, municipal wastewater, and the Rio Grande in New Mexico. Sci Total Environ 2006;366:772–83. Cermola M, DellaGreca M, Iesce MR, Previtera L, Rubino M, Temussi F, et al. Phototransformation of fibrate drugs in aqueous media. Environ Chem Lett 2005;3:43–7. Cole DW, Cole R, Gaydos SJ, Gray J, Hyland G, Jacques ML, et al. Aquaculture: environmental, toxicological, and health issues. Int J Hyg Environ Health 2009;212: 369–77. Commission Decision (2002/657/EC) of 12 August 2002, Implementing Council Directive 96/23/EC concerning the performance of analytical methods and the interpretation of results, Official Journal of the European Communities L221, Brussels, Belgium, pp. 8–36. Conley JM, Symes SJ, Schorr MS, Richards SM. Spatial and temporal analysis of pharmaceutical concentrations in the upper Tennessee River basin. Chemosphere 2008;73:1178–87. Cortes R. Rio Douro: caracterização ecológica e valores ambientais. As águas do Douro. Águas do Douro e Paiva; Edições Afrontamento, 2009, pp. 201–214. Daneshvar A, Svanfelt J, Kronberg L, Prévost M, Weyhenmeyer GA. Seasonal variations in the occurrence and fate of basic and neutral pharmaceuticals in a Swedish river– lake system. Chemosphere 2010;80:301–9. Daughton CG, Ternes TA. Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ Health Perspect 1999;107:907–38. Díaz-Cruz MS, García-Galán MJ, Barceló D. Highly sensitive simultaneous determination of sulfonamide antibiotics and one metabolite in environmental waters by liquid chromatography-quadrupole linear ion trap-mass spectrometry. J Chromatogr A 2008;1193:50–9. El-Gindy A, Emara S, Mesbah MK, Hadad GM. Spectrophotometric and liquid chromatographic determination of fenofibrate and vinpocetine and their hydrolysis products. Farmaco 2005;60:425–38. Feitosa-Felizzola J, Chiron S. Occurrence and distribution of selected antibiotics in a small Mediterranean stream (Arc River, Southern France). J Hydrol 2009;364:50–7. Fent K, Weston AA, Caminada D. Ecotoxicology of human pharmaceuticals. Aquat Toxicol 2006;76:122–59. Ferreira M, Antunes P, Gil O, Vale C, Reis-Henriques MA. Organochlorine contaminants in flounder (Platichthys flesus) and mullet (Mugil cephalus) from Douro estuary, and their use as sentinel species for environmental monitoring. Aquat Toxicol 2004;69: 347–57. Ferreira M, Moradas-Ferreira P, Reis-Henriques MA. The effect of long-term depuration on phase I and phase II biotransformation in mullets (Mugil cephalus) chronically exposed to pollutants in River Douro estuary, Portugal. Mar Environ Res 2006;61: 326–38. Gros M, Petrovic M, Ginebreda A, Barceló D. Removal of pharmaceuticals during wastewater treatment and environmental risk assessment using hazard indexes. Environ Int 2010;36:15–26. Heberer T. Tracking persistent pharmaceutical residues from municipal sewage to drinking water. J Hydrol 2002;266:175–89. Hirsch R, Ternes T, Haberer K, Kratz K-L. Occurrence of antibiotics in the aquatic environment. Sci Total Environ 1999;225:109–18. Huggett DB, Brooks BW, Peterson B, Foran CM, Schlenk D. Toxicity of select beta adrenergic receptor-blocking pharmaceuticals (b-blockers) on aquatic organisms. Arch Environ Contam Toxicol 2002;43:229–35. INFARMED. Autoridade Nacional do Medicamento e Produtos de Saúde, I.P; 2009 (http: // (accessed may). Isidori M, Nardelli A, Pascarella L, Rubino M, Parrella A. Toxic and genotoxic impact of fibrates and their photoproducts on non-target organisms. Environ Int 2007;33: 635–41. Kasprzyk-Hordern B, Dinsdale RM, Guwy AJ. Multi-residue method for the determination of basic/neutral pharmaceuticals and illicit drugs in surface water by solidphase extraction and ultra performance liquid chromatography-positive electrospray ionisation tandem mass spectrometry. J Chromatogr A 2007;1161:132–45. Khetan SK, Collins TJ. Human pharmaceuticals in the aquatic environment: a challenge to green chemistry. Chem Rev 2007;107:2319–64. Kümmerer K. The presence of pharmaceuticals in the environment due to human use — present knowledge and future challenges. J Environ Manage 2009;90:2354–66. Langford KH, Thomas KV. Determination of pharmaceutical compounds in hospital effluents and their contribution to wastewater treatment works. Environ Int 2009;35:766–70. Larsson DGJ, de Pedro C, Paxeus N. Effluent from drug manufactures contains extremely high levels of pharmaceuticals. J Hazard Mater 2007;148:751–5. Li B, Zhang T. Biodegradation and adsorption of antibiotics in the activated sludge process. Environ Sci Technol 2010;44:3468–73. Loos R, Gawlik BM, Locoro G, Rimaviciute E, Contini S, Bidoglio G. EU-wide survey of polar organic persistent pollutants in European river waters. Environ Pollut 2009;157:561–8. Madureira TV, Barreiro JC, Rocha MJ, Cass QB, Tiritan ME. Pharmaceutical trace analysis in aqueous environmental matrices by liquid chromatography-ion trap tandem mass spectrometry. J Chromatogr A 2009;1216:7033–42.


T.V. Madureira et al. / Science of the Total Environment 408 (2010) 5513–5520

Madureira TV, Rocha MJ, Cass QB, Tiritan ME. Development and optimization of a HPLCDAD method for the determination of diverse pharmaceuticals in estuarine surface waters. J Chromatogr Sci 2010;48:176–82. Magalhães C, Teixeira C, Teixeira R, Machado A, Azevedo I, Bordalo AA. Dissolved organic carbon and nitrogen dynamics in the Douro River estuary, Portugal. Cienc Mar 2008;34:271–82. Miao XS, Metcalfe CD. Determination of carbamazepine and its metabolites in aqueous samples using liquid chromatography-electrospray tandem mass spectrometry. Anal Chem 2003;75:3731–8. Moldovan Z. Occurrences of pharmaceutical and personal care products as micropollutants in rivers from Romania. Chemosphere 2006;64:1808–17. Mompelat S, Le Bot B, Thomas O. Occurrence and fate of pharmaceutical products and by-products, from resource to drinking water. Environ Int 2009;35:803–14. Morley NJ. Environmental risk and toxicology of human and veterinary waste pharmaceutical exposure to wild aquatic host–parasite relationships. Environ Toxicol Pharmacol 2009;27:161–75. Mucha AP, Vasconcelos MTSD, Bordalo AA. Spatial and seasonal variations of the macrobenthic community and metal contamination in the Douro estuary (Portugal). Mar Environ Res 2005;60:531–50. Nakada N, Kiri K, Shinohara H, Harada A, Kuroda K, Takizawa S, et al. Evaluation of pharmaceuticals and personal care products as water-soluble molecular markers of sewage. Environ Sci Technol 2008;42:6347–53. Öllers S, Singer HP, Fässler P, Müller SR. Simultaneous quantification of neutral and acidic pharmaceuticals and pesticides at the low-ng/l level in surface and waste water. J Chromatogr A 2001;911:225–34. Rath NP, Haq W, Balendiran GK. Fenofibric acid. Acta Crystallogr, Sect C: Cryst Struct Commun 2005;61:o81–4. Ribeiro C, Tiritan ME, Rocha E, Rocha MJ. Seasonal and spatial distribution of several endocrine-disrupting compounds in the Douro River estuary, Portugal. Arch Environ Contam Toxicol 2009;56:1-11. Roberts PH, Thomas KV. The occurrence of selected pharmaceuticals in wastewater effluent and surface waters of the lower Tyne catchment. Sci Total Environ 2006;356:143–53. Sacher F, Ehmann M, Gabriel S, Graf C, Brauch H-J. Pharmaceutical residues in the river Rhine — results of a one-decade monitoring programme. J Environ Monit 2008;10: 664–70. Santos JL, Aparicio I, Callejón M, Alonso E. Occurrence of pharmaceutically active compounds during 1-year period in wastewaters from four wastewater treatment plants in Seville (Spain). J Hazard Mater 2009;164:1509–16.

Santos LHMLM, Araújo AN, Fachini A, Pena A, Delerue-Matos C, Montenegro MCBSM. Ecotoxicological aspects related to the presence of pharmaceuticals in the aquatic environment. J Hazard Mater 2010;175:45–95. Sirtori C, Agüera A, Gernjak W, Malato S. Effect of water-matrix composition on trimethoprim solar photodegradation kinetics and pathways. Water Res 2010;44: 2735–44. Tamtam F, Mercier F, Le Bot B, Eurin J, Tuc Dinh Q, Clement M, et al. Occurrence and fate of antibiotics in the Seine River in various hydrological conditions. Sci Total Environ 2008;393:84–95. Ternes TA. Occurrence of drugs in German sewage treatment plants and rivers. Water Res 1998;32:3245–60. Ternes TA, Herrmann N, Bonerz M, Knacker T, Siegrist H, Joss A. A rapid method to measure the solid-water distribution coefficient (Kd) for pharmaceuticals and musk fragrances in sewage sludge. Water Res 2004;38:4075–84. Topp E, Monteiro SC, Beck A, Coelho BB, Boxall ABA, Duenk PW, et al. Runoff of pharmaceuticals and personal care products following application of biosolids to an agricultural field. Sci Total Environ 2008;396:52–9. Triebskorn R, Casper H, Scheil V, Schwaiger J. Ultrastructural effects of pharmaceuticals (carbamazepine, clofibric acid, metoprolol, diclofenac) in rainbow trout (Oncorhynchus mykiss) and common carp (Cyprinus carpio). Anal Bioanal Chem 2007;387: 1405–16. Trovó AG, Nogueira RFP, Agüera A, Sirtori C, Fernández-Alba AR. Photodegradation of sulfamethoxazole in various aqueous media: persistence, toxicity and photoproducts assessment. Chemosphere 2009;77:1292–8. Vieno NM, Tuhkanen T, Kronberg L. Analysis of neutral and basic pharmaceuticals in sewage treatment plants and in recipient rivers using solid phase extraction and liquid chromatography-tandem mass spectrometry detection. J Chromatogr A 2006;1134:101–11. Williams RT. Human pharmaceuticals: assessing the impacts on aquatic ecosystems. Society of Environmental Toxicology and Chemistry (SETAC); 2003. Xia K, Bhandari A, Das K, Pillar G. Occurrence and Fate of Pharmaceuticals and Personal Care Products (PPCPs) in biosolids. J Environ Qual 2005;34:91-104. Yamamoto H, Nakamura Y, Moriguchi S, Nakamura Y, Honda Y, Tamura I, et al. Persistence and partitioning of eight selected pharmaceuticals in the aquatic environment: laboratory photolysis, biodegradation, and sorption experiments. Water Res 2009;43:351–62. Zhang ZL, Zhou JL. Simultaneous determination of various pharmaceutical compounds in water by solid-phase extraction-liquid chromatography-tandem mass spectrometry. J Chromatogr A 2007;1154:205–13.