SEARCH

SEARCH BY CITATION

Keywords:

  • 17α-Ethinylestradiol;
  • Estrogen;
  • Surface water;
  • Water quality model

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. SUPPORTING INFORMATION
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

An evaluation of measured and predicted concentrations of 17α-ethinylestradiol in surface waters of the United States and Europe was conducted to develop expected long-term exposure concentrations for this compound. Measured environmental concentrations (MECs) in surface waters were identified from the literature. Predicted environmental concentrations (PECs) were generated for European and U.S. watersheds using the GREAT-ER and PhATE™ models, respectively. The majority of MECs are nondetect and generally consistent with model PECs and conservative mass balance calculations. However, the highest MECs are not consistent with concentrations derived from conservative (worst-case) mass balance estimates or model PECs. A review of analytical methods suggests that tandem or high-resolution mass spectrometry methods with extract cleanup result in lower detection limits and lower reported concentrations consistent with model predictions and bounding estimates. Based on model results using PhATE and GREAT-ER, the 90th-percentile low-flow PECs in surface water are approximately 0.2 and 0.3 ng/L for the United States and Europe, respectively. These levels represent conservative estimates of long-term exposure that can be used for risk assessment purposes. Our analysis also indicates that average concentrations are one to two orders of magnitude lower than these 90th-percentile estimates. Higher reported concentrations (e.g., greater than the 99th-percentile PEC of ∼1 ng/L) could result from methodological problems or unusual environmental circumstances; however, such concentrations are not representative of levels generally found in the environment, warrant special scrutiny, and are not appropriate for use in risk assessments of long-term exposures.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. SUPPORTING INFORMATION
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

The occurrence of 17α-ethinylestradiol (EE2) in surface water has been the subject of many field investigations with reported concentrations ranging from below detection limits of 0.01 ng/L [1] to detections as high as 273 ng/L [2]. The objective of the present study is to derive appropriate exposure concentrations for use in human and aquatic life risk assessments. This analysis evaluates concentrations in surface water resulting from human use and excretion, which is the predominant pathway for the introduction of EE2 into the environment. Other possible sources of EE2 in the environment, such as off-label use for veterinary purposes, have not been reported and therefore are not considered. This analysis is an exposure assessment only. A risk assessment for EE2 is the subject of a future paper.

The most reliable methods to detect and quantify EE2 in aqueous matrices are gas chromatography/tandem mass spectrometry (GC-MS/MS), liquid chromatography/tandem mass spectrometry (LC-MS/MS) [3], and high-resolution mass spectrometry [4–7]. Single MS measurement can lead to an overestimation of the concentration of EE2 due to an overlap of the EE2 peak in the chromatogram with an unknown impurity of similar mass-to-charge ratio [3,8]. The results of an international round-robin analysis of several estrogenic hormones, including EE2, also suggest that tandem MS with deuterated or 13C-labeled internal standards and extract cleanup is preferred for EE2 analysis [9 (www-32.cis.portlandcs.net/wpt/001/0033/0010033.pdf), 10].

This investigation evaluates measured concentrations reported in the literature, simple mass balance calculations, and predicted concentrations from georeferenced water quality models to derive concentrations of EE2 in surface water for the purpose of evaluating potential risks across a watershed or nation. The use of both measured data and model predictions has been recommended as a robust approach to risk assessment of chemicals in surface water [11].

The synthetic contraceptive agent EE2 is typically administered to patients in doses of 20 to 35 μd for 21 d out of each 28-d cycle [12]. In most countries, annual prescription volumes total approximately 100 kg/year or less (IMS Health, IMS MIDAS Quantum, Year End, March 2007). The high endocrine potency of EE2 has generated many research studies in the laboratory and the field to determine its fate and effects in the environment.

To keep the present study as concise as possible, relevant background information (Supporting Information; http://dx.doi.org/10.1897/08–622.S1) is provided via the Internet.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. SUPPORTING INFORMATION
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Compilation of measured environmental concentrations

Measured environmental concentrations (MECs) of EE2 in surface waters were compiled from the peer-reviewed English-language literature. All available papers published through the end of 2006 were reviewed. Some papers present only summary statistics of measured concentrations. The authors of these papers were contacted and asked to provide a complete data set for use in this evaluation. Some authors did not respond to this request, in which case all the available information in the published paper (e.g., number of samples, number of detects, minimum, average, and maximum concentrations) was used to approximate their data sets for this evaluation. In addition, a synoptic review of the methods employed by these investigators (e.g., sample collection and preservation, sample preparation, instrumentation, detection methods, and detection limits) was conducted. Measured environmental concentrations and associated methods were compiled in a relational database that was used to explore potential relationships among methods and reported concentrations.

Notably, reported MECs do not represent random assessments of potential exposures of EE2 in surface waters. In general, sampling schemes were devised to target areas suspected of containing EE2, such as downstream of sewage treatment plants (STPs). Thus, MECs represent the distribution of concentrations in impacted areas rather than their overall distribution across entire watersheds.

Predicted environmental concentrations in surface waters

Predicted environmental concentrations (PECs) for EE2 in North American and European surface waters were estimated using the PhATE™ version 2.1 [13] and GREAT-ER desktop version 2.0.0 [14] models, respectively. These models are used to estimate surface water concentrations across a wider range of environmental conditions than is practicable with field measurements. In addition, the models can predict surface water concentrations that are below the analytical limits of detection for EE2. Descriptions of these models and their availability are provided elsewhere [13–15].

PhATE was used to estimate concentrations in stream segments throughout 11 watersheds in the United States [13]. GREAT-ER was used to estimate concentrations in stream segments throughout five watersheds in Germany, Italy, and the United Kingdom [16; http://www.great-er.org/pages/Catchments.cfm). PhATE incorporates only those stream segments that are located below an STP (i.e., segments potentially impacted by EE2 in STP effluents). By contrast, watersheds incorporated in GREAT-ER contain a significant number of segments (45–94% depending on the watershed) that are nonimpacted (i.e., PEC equals zero). For this assessment, only nonzero values were used in the data analyses as a conservative approach and to make comparisons of PECs with MECs on a similar basis (i.e., evaluating only potentially affected surface waters).

PhATE was run under both mean and 7Q10 low-flow (i.e., the lowest consecutive 7-d flow that occurs on average once every 10 years) conditions. The 90th-percentile Csim from GREAT-ER represents the 90th-percentile PEC for a particular segment from a series of Monte Carlo simulations. The 90th percentile of the 90th-percentile Csim concentrations for all the combined nonzero segments for the five watersheds in GREAT-ER was used to estimate low-flow concentrations in Europe, and the 50th percentile of the 50th-percentile Csim concentrations was used to estimate mean flow concentrations. Use of the 90th-percentile value of the low-flow predicted concentrations from PhATE and GREAT-ER is considered to provide a conservative estimate of exposure for risk assessment purposes [17–20].

Model quality control runs using a conservative tracer compound indicated that PECs in certain segments from several GREAT-ER catchments were higher than concentrations that would be observed in untreated STP effluent based on per capita sewage flows provided in the model. These outlier segments represent a small percentage of segments in each catchment (0.0–3.5%). Calculated errors for the tracer compound are generally minor (less than a factor of two); however, larger errors were observed in a small number of segments. Five catchments with no or only minor outlier segments were used for this analysis: Aire, Don-Rother, Itter, Lambro, and Went. Predicted environmental concentrations from the Elbe, Mayenne, Rhine, Rupel, and Rur catchments were excluded because they contain outlier segments where PECs exceed the maximum expected concentration by a factor of two or greater. Details of the outlier segments are presented in the Supporting Information (http://dx.doi.org/10.1897/08–622.S1).

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. SUPPORTING INFORMATION
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Introduction of EE2 into wastewater

Estimates of the total annual mass of EE2 sold in Belgium, France, Germany, Italy, The Netherlands, the United Kingdom, and the United States for the 12-month period ending March 2007 were obtained from IMS Health (IMS MIDAS Quantum, Year End, March 2007). Annual sales volumes ranged from 850 ng/capita/d in the United States to 2,590 ng/capita/d in The Netherlands (Table 1). The annual use of EE2 reflects sales of all products containing EE2 (including combination products) to retail pharmacies, nonfederal hospitals, federal facilities, long-term care facilities, clinics, and health maintenance organizations. The annual mass was not adjusted for drug products that were sold but not used.

An indirect estimate of the annual mass of EE2 used in the United States is obtained using demographic data and the average daily therapeutic dose of EE2 (26.25 μd; 35 μtablet × 21 tablets/28 d). Approximately 11.6 million women in the United States (8.1% of all women in the United States) used EE2 for contraception in 2002 [21]. In 2007, the population of women in the United States increased to approximately 155 million. Assuming that the percent of women using EE2 has remained the same since 2002, the mass of EE2 used in 2007 would be approximately 120 kg/year. This estimate is 36% greater than the annual mass of EE2 obtained from IMS Health (Table 1). The sensitivity of model PECs to this range in annual usage (i.e., 88–120 kg) is presented in a later section.

Johnson and Williams [22] evaluated the fraction of the EE2 dose excreted based on measurements in urine and feces. They estimated that the fraction of EE2 excreted in feces is 30%, of which 77% is present as EE2, or 23% of the dose, and that the fraction excreted in urine is 27%, including glucoronide and sulfate conjugates. Although Johnson and Williams assumed that sulfate conjugates are not deconjugated, there is some evidence of sulfate deconjugation in humans, so this analysis assumes that deconjugation also occurs in STPs to provide a conservative (i.e., higher) estimate of EE2 concentrations. Therefore, for the purposes of this evaluation, it is assumed that 50% of the EE2 dose (23% in feces plus 27% in urine) is excreted by patients as EE2 or as conjugates subject to deconjugation in STPs.

Table Table 1.. 17α-ethinylestradiol (EE2) usage by country for the 12-month period ending March 2007 and sewage treatment plant (STP) effluent concentrations of EE2 derived from various assumptions regarding metabolism and STP removal efficiency and assuming a water use of 200 L/capita/d
    STP effluent concn., ng/L
  Patient use   
CountryModel populationkg/yearang/capita/dNo metabolism, no treatment50% metabolism, no treatment50% metabolism, 82% secondary removal
  1. a IMS Health, IMS MIDAS Quantum, Year End, March 2007.

Belgium10,400,0008.02,11010.65.30.9
France59,900,00033.61,5407.73.90.7
Germany82,500,00051.01,6908.54.20.8
Italy57,900,00019.99404.72.40.4
Netherlands16,300,00015.42,59013.06.51.2
United Kingdom59,700,00026.31,2106.13.00.5
United States285,000,00088.08504.32.10.4
thumbnail image

Figure Fig. 1.. Cumulative probability distribution of measured 17α-ethinylestradiol concentrations (n = 1,652) in surface waters. Nondetect values (diamond shapes, ♦) are plotted at the reported limit of detection in ascending order followed by the detected values (square shapes, □) plotted in ascending order. STP = sewage treatment plant; LOD = limit of detection.

Download figure to PowerPoint

Fate of EE2 in STPs and surface water

The average removal efficiency of EE2 in primary STPs was estimated to be 10% using EpiSuite 3.02 (Syracuse Research Corporation). The removal efficiency in secondary STPs (e.g., activated sludge and trickling filters) was estimated to be 82%, an average of measured values [23–29]. An in-stream depletion rate of zero was used for the model runs as a conservative assumption in light of limited available data to estimate this parameter.

thumbnail image

Figure Fig. 2.. Cumulative probability distribution of measured 17α-ethinylestradiol concentrations (n = 360) in surface waters using gas chromatography/tandem mass spectrometry or liquid chromatography/tandem mass spectrometry detection with extract cleanup following extraction. STP = sewage treatment plant; LOD = limit of detection.

Download figure to PowerPoint

Mass balance calculation in STP effluent

Expected concentrations of EE2 in STP effluent are calculated using a mass balance approach based on estimates of per capita use of EE2, a conservative estimate of per capita water use (200 L/capita/d), loss of EE2 via metabolism, and loss via removal in treatment (Table 1). These assumptions result in a maximum STP effluent concentration ranging from approximately 13 ng/L (see estimate for The Netherlands with no metabolism and no STP removal) to 0.4 ng/L (see estimate for the United States and Italy assuming 50% human metabolism and 82% STP removal). Concentrations in surface water would be lower as a result of in-stream dilution and instream degradation; however, these estimates are used as an upper-bound point of comparison to assess the feasibility of measured concentrations.

Compilation of MECs in surface water and associated analytical methods

A literature search uncovered 52 papers that report EE2 concentrations in surface water in 16 countries (see Supporting Information for list of references). These papers span a range of analytical detection methods from immunoassay to MS to tandem MS methods. Individual MECs from these papers were compiled into a database along with details regarding sample location and analytical methods.

A cumulative probability distribution of all MECs (n = 1,652) is presented in Figure 1. Concentrations range from nondetect to 273 ng/L. The 90th-percentile concentration is 1.7 ng/L. Approximately 70% of the measurements are nondetect with limits of detection ranging from 0.01 to 30 ng/L. Note that 24 of 1,652 reported MECs (1.5%) in surface water are greater than the maximum expected STP effluent concentration of 13 ng/L (Fig. 1). Anderson et al. [13] evaluated some of the highest MECs reported in surface water (i.e., 73 and 273 ng/L) and concluded that they are unlikely to result from human use.

For samples analyzed by GC-MS/MS or LC-MS/MS with an additional cleanup step following the extraction (n = 360), 87% of measurements are nondetect, with limits of detection ranging from 0.1 to 1 ng/L; concentrations range from nondetect to 4.6 ng/L; and the 90th-percentile concentration is 0.43 ng/L (Fig. 2). A similar range of concentrations (nondetect to 5.1 ng/L; n = 35) was also found without the additional cleanup step when high-resolution MS methods were employed. None of the MECs analyzed using tandem MS methods (either with or without extract cleanup) or high-resolution MS methods exceed the maximum STP effluent concentration of 13 ng/L derived from mass balance estimates.

thumbnail image

Figure Fig. 3.. Cumulative probability distribution of predicted concentrations of 17α-ethinylestradiol in surface water for 11 U.S. watersheds using the PhATE™ (Washington, DC, USA) model. PEC = predicted environmental concentration.

Download figure to PowerPoint

As a result of the data and method analyses, two findings are reported. First, single MS methods can overestimate EE2 concentrations as compared with high-resolution or tandem MS methods. This is substantiated by an international round-robin test in which eight laboratories reported results for EE2 in wastewater. Single-stage MS analysis of EE2 in wastewater resulted in an outlier value of 50 ng/L, and single-stage LC/MS resulted in a slightly overestimated value of 3.4 ng/L compared to the mean value of 1.38 ng/L obtained by four laboratories using tandem MS methods [9]. The other two laboratories did not detect EE2 in the sample. Second, while the use of an isotopic EE2 internal standard is a preferred approach to account for suppression of ionization and recovery, it alone cannot compensate for an interference or reduce the likelihood of generating a high MEC, as seen with 79 of the 86 (92%) reported high MECs (<5.0 ng/L), all determined from single quad MS methods using an isotopic EE2 internal standard. The previous two findings, along with the recommendations of Johnson et al. [11], suggest that an ideal method would ensure specificity by employing either high-resolution MS or tandem MS methods in conjunction with sample cleanup procedures and using three or more ions for quantification, additional accuracy and reproducibility by using an isotopic internal standard, appropriate validation at or near environmental levels of EE2, and that samples are assayed as soon as possible after collection.

Predicted concentrations in surface water

Cumulative probability distributions of PECs from PhATE and GREAT-ER are presented in Figures 3 and 4, respectively. For PhATE under mean flow conditions, the median, 90th-percentile, and maximum PECs are 0.00064, 0.0075, and 0.46 ng/L, respectively, while under low-flow conditions, the median, 90th-percentile, and maximum PECs are 0.0063, 0.16, and 0.58 ng/L, respectively. For GREAT-ER under mean flow conditions, the median, 90th-percentile, and maximum Csim values are 0.054, 0.13, and 0.62, respectively, while under low-flow conditions, the median, 90th-percentile, and maximum Csim values are 0.10, 0.25, and 1.3 ng/L, respectively. Predicted concentrations of EE2 in surface water from PhATE and GREAT-ER are consistent with mass balance calculations. The maximum PEC for EE2 from GREAT-ER is 1.3 ng/L, which is consistent with the mass balance estimates of 0.4 to 1.2 ng/L for Europe based on secondary treatment (Table 1). The maximum PEC for EE2 from PhATE is 0.58 ng/L, which is consistent with the mass balance estimate of 0.38 ng/L for the United States based on secondary treatment (Table 1, rounded to 0.4 ng/L). These high surface water concentrations are likely to occur in effluent-dominated stream segments with little or no in-stream dilution.

Predicted environmental concentrations from GREAT-ER are higher than those from PhATE. The ratio of GREAT-ER to PhATE PEC under low-flow conditions is 1.6 and 2.2 for the 90th-percentile and maximum PECs, respectively, and the ratio under mean flow conditions is 17 and 1.4 for the 90th-percentile and maximum PECs, respectively. The higher surface water PECs from GREAT-ER are a result of higher per capita use of EE2, lower per capita water use, and lower in-stream dilution in GREAT-ER watersheds as compared with PhATE. The average per capita water use in the PhATE watersheds is approximately 50% higher than in the GREATER watersheds evaluated (average of 580 L/capita/d in PhATE vs 390 L/capita/d in GREAT-ER). The average per capita use of EE2 in the GREAT-ER watersheds is approximately 50% higher than in the PhATE watersheds (average of 1,280 ng/capita/d in GREAT-ER vs 850 ng/capita/d in PhATE). These two factors account for most of the difference between the 90th-percentile PECs from GREAT-ER and PhATE under low-flow conditions where dilution would be less significant as compared with mean flow conditions. The larger difference between the 90th-percentile PECs under mean flow conditions is likely due to greater in-stream dilution in the PhATE watersheds as compared with GREAT-ER.

thumbnail image

Figure Fig. 4.. Cumulative probability distribution of predicted concentrations of 17α-ethinylestradiol in surface water for five European watersheds using the GREAT-ER model (Brussels, Belgium). PEC = predicted environmental concentration.

Download figure to PowerPoint

As discussed previously, the per capita use of EE2 is a key input model parameter, and model PECs are directly proportional to this parameter. An indirect estimate of EE2 use in the United States based on demographic information of 120 kg/year was presented, which is 36% greater than the use obtained from IMS Health (IMS MIDAS Quantum, Year End, March 2007). If the demographic estimate of EE2 usage was used in the PhATE model analysis, then model PECs would be 36% greater than those presented previously. At low flow, the 90th-percentile PEC would increase from 0.16 to 0.22 ng/L, and the maximum PEC would increase from 0.58 to 0.80 ng/L.

Bounding estimates of EE2 concentrations in wastewater and surface water

The bounding estimates presented here assume that human therapeutic use is the only source of EE2. This assumption is based on discussions with veterinarians and animal control experts, none of whom identified or knew of nonhuman uses of EE2.

The therapeutic dosing regimen for EE2 is used to develop a range of maximum possible STP influent concentrations in the United States. Progressively more realistic wastewater concentrations are estimated by accounting for U.S. water use and demographic patterns of EE2 use, and these estimates for the United States are then extended to Europe. For this calculation, a woman is assumed to ingest 35 μd of EE2 (the dose of EE2 in a tablet) during 21 d out of each 28-d cycle [12]. A conservatively low estimate of water use in the United States is 200 L/capita/d, resulting in an EE2 wastewater concentration of 175 ng/L for a population composed entirely of women using EE2. Accounting for metabolism, the concentration of EE2 in wastewater from such a population would be 88 ng/L. As previously discussed, approximately 8.1% of women (or 4.1% of the total population) in the United States use EE2 as a contraceptive. Therefore, the concentration of EE2 in the influent of an STP serving a typical U.S. community is expected to be approximately 4 ng/L (i.e., 4.1% of 88 ng/L). Note that these estimates assume a daily EE2 dose of 35 μd; however, the average daily dose over each 28-d cycle is 26.25 μd [12]. Thus, the typical concentration of EE2 in STP influent in the United States would be approximately 3 ng/L. Following the same calculation for Europe, with their higher EE2 consumption and relatively lower water use, no values greater than 9 ng/L would be expected in European STP influent.

It follows that even assuming no removal in the STP and no dilution in the receiving water, long-term EE2 concentrations in typical surface waters are unlikely to exceed 3 ng/L in the United States or 9 ng/L in Europe. Given that STPs are expected to remove some EE2 and because some dilution and in-stream depletion occurs in most receiving waters most of the time, almost all surface waters would be expected to have considerably lower EE2 concentrations. This conclusion is consistent with predicted concentrations from the PhATE and GREAT-ER models: maximum low-flow PECs are 0.58 ng/L and 1.3 ng/L for the United States and Europe, respectively, while 90th-percentile low-flow PECs are 0.16 ng/L and 0.25 ng/L for the United States and Europe, respectively (Figs. 3 and 4). Although mass balance calculations can provide a good first estimate of maximum concentrations of EE2 expected in surface water, georeferenced water quality models are more appropriate for evaluating long-term exposures across an entire watershed under a wide range of stream flows and STP inputs [11].

These bounding calculations do not mean that higher concentrations in STP influents and surface waters are impossible. As described previously, higher concentrations could occur under special circumstances. However, such circumstances are likely to be isolated and rare and are unlikely to account for the numerous reported EE2 surface water concentrations (260 of 1,652 MECs, or ∼16%; Fig. 1) that are greater than the 99th-percentile PEC of approximately 1 ng/L (Figs. 3 and 4). Thus, reported concentrations in surface waters exceeding approximately 1 ng/L warrant special scrutiny and detailed investigation to evaluate unusual circumstances leading to the occurrence or detection of EE2 at these elevated levels. Some of the highest concentrations reported in surface water are likely erroneous values caused by analytical interferences as discussed previously.

The approach presented here for evaluating exposure concentrations of EE2 using both measured concentrations and model predictions is also relevant for exposure assessments of other pharmaceutical compounds. Field measurements are essential for model corroboration and developing accurate model input parameters. Georeferenced models such as PhATE and GREAT-ER can be used for designing field sampling programs, for cost-effective screening to identify compounds of potential concern, and to establish environmentally relevant concentrations for aquatic toxicity testing. Watershed models can be used to evaluate exposure concentrations over a wider range of conditions (e.g., flow, annual use, and treatment efficiency) than is practical with field sampling programs. As is shown in this assessment for EE2, model predictions can also help identify measurement outliers that may not be representative of long-term exposures.

SUPPORTING INFORMATION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. SUPPORTING INFORMATION
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Fig. S1. Chemical structure of EE2.

Table S1. PhATE and GREAT-ER model input parameters for EE2.

Table S2. Results of analysis for a conservative tracer compound with a use of 54.75 mg/capita/year. The number of outlier segments where the PEC exceeds 1,000 ng/L is shown for each catchment in GREAT-ER.

All found at DOI: 10.1897/08–622.S1 (118 KB PDF).

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. SUPPORTING INFORMATION
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

The authors acknowledge with appreciation the significant contribution of Pharmaceutical Research and Manufacturers of America (PhRMA) in supporting this work. The authors are grateful to the following investigators for providing supporting data on measured concentrations: P. Hohenblum, Umweltbundesamt, Vienna, Austria; L. Shore, Kimron Veterinary Institute, Israel; and R. Williams, Centre for Ecology and Hydrology, Oxfordshire, UK. Identification of published articles that report surface water concentrations for EE2 was greatly facilitated by a database maintained by R.T. Williams, Pfizer. The authors also acknowledge the contributions of D. Pfeiffer and B. Duplessie of AMEC Earth&Environmental.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. SUPPORTING INFORMATION
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information
  • 1
    Hintermann T, Schneider C, Scholer HF, Schneider RJ. 2006. Field study using two immunoassays for the determination of estradiol and ethinylestradiol in the aquatic environment. Water Res 40: 22872294.
  • 2
    Kolpin DW, Furlong ET, Meyer MT, Thurman EM, Zaugg SD, Barber LB, Buxton HT. 2002. Pharmaceuticals, hormones, and other organic wastewater contaminant in U.S. streams, 1999–2000: A national reconnaissance. Environ Sci Technol 36: 12021211.
  • 3
    Ternes TA, Kreckel P, Muller J. 1999. Behaviour and occurrence of oestrogens in municiple sewage treatment plants—II. Aerobic batch experiments with activated sludge. Sci Total Environ 225: 9199.
  • 4
    U.S. Environmental Protection Agency. 2007. Steroids and hormones in water, soil, sediment, and biosolids by HRGC/HRMS. Method 1698. EPA-821-R-08–003. Washington, DC.
  • 5
    Aerni HR, Kobler B, Rutishauser BV, Wettstein FE, Fischer R, Giger W, Hungerbuhler A, Marazuela MD, Peter A, Schonenberger R, Vogeli AC, Suter MJF, Eggen RIL. 2004. Combined biological and chemical assessment of estrogenic activities in wastewater treatment plant effluents. Anal Bioanal Chem 378: 688696.
  • 6
    Andersson J, Woldegiorgis A, Remberger M, Kaj L, Ekheden Y, Dusan B, Svenson A, Brorstrom-Lunden E, Dye C, Schlabach M. 2006. Results from the Swedish National Screening Programme 2005, Subreport 1: Antibiotics, Anti-inflammatory Substances, and Hormones. B1689. Swedish Environmental Research Institute, Stockholm.
  • 7
    Kuch HM, Ballschmiter K. 2001. Determination of endocrine-disrupting phenolic compounds and estrogens in surface and drinking water by HRGC-(NCI)-MS in the picogram per liter range. Environ Sci Technol 35: 32013206.
  • 8
    Huang CH, Sedlak DL. 2001. Analysis of estrogenic hormones in municipal wastewater effluent and surface water using enzyme-linked immunosorbent assay and gas chromatography/tandem mass spectrometry. Environ Toxicol Chem 20: 133139.
  • 9
    Esperanza M, Herry G, Manciot F, Laǐné JM. 2006. Analysis of estrogenic hormones in natural waters, wastewater and sludge: Results from the First International Round Robin Test. Water Practice and Technology.
  • 10
    Richardson SD, Ternes TA. 2005. Water analysis: Emerging contaminants and current issues. Anal Chem 77: 38073838.
  • 11
    Johnson A, Ternes TA, Williams R, Sumpter JP. 2008. Assessing the concentrations of polar organic microcontaminants from point sources in the aquatic environment: Measure or model?. Environ Sci Technol 42: 53905399.
  • 12
    Physician's Desk Reference. 2007. PDR Electronic Library. Thomson MICROMEDEX, Montvale, NJ, USA.
  • 13
    Anderson PD, D'Aco VJ, Shanahan P, Chapra SC, Buzby ME, Cunningham VL, Duplessie BM, Hayes EP, Mastrocco FJ, Parke NJ, Rader JC, Samuelian JH, Schwab BW. 2004. Screening analysis of human pharmaceutical compounds in U.S. surface waters. Environ Sci Technol 38: 838849.
  • 14
    Feijtel TCJ, Boeije G, Matthies M, Young A, Morris G, Gandolfi C, Hansen B, Fox K, Holt M, Koch V, Schroder R, Cassani G, Schowanek D, Rosenblom J, Niessen H. 1997. Development of a geography-referenced regional exposure assessment tool for European Rivers—GREAT-ER. Contribution to GREAT-ER 1. Chemosphere 34: 23512373.
  • 15
    Cunningham VL, Buzby ME, Hutchinson T, Mastrocco FJ, Parke NJ, Roden NM. 2006. Effects of human pharmaceuticals on aquatic life: Next steps. Environ Sci Technol 40: 34563462.
  • 16
    European Chemical Industry Long-Range Research Initiative. 2007. A GIS assisted model for environmental risk assessment and management of chemicals in river basins. Brussels, Belgium.
  • 17
    Fox KK, Daniel M, Morris G, Holt M. 2000. The use of measured boron concentration data from the GREAT-ER UK validation study (1996–1998) to generate predicted regional boron concentrations. Sci Total Environ 251/ 252: 305316.
  • 18
    Rapaport RA. 1988. Prediction of consumer product chemical concentrations as a function of publicly owned treatment works treatment type and river dilution. Environ Toxicol Chem 7: 107115.
  • 19
    U.S. Environmental Protection Agency. 1992. Guidelines for Exposure Assessment. EPA/600/Z-92/001. Washington, DC.
  • 20
    Versteeg DJ, Alder AC, Cunningham VL, Kolpin DW, Murray-Smith R, Ternes TA. 2005. Environmental exposure modeling and monitoring of human pharmaceuticals in the environment. In WilliamsRT, ed, Science for Assessing the Impacts of Human Pharmaceuticals on Aquatic Ecosystems. SETAC, Pensacola, FL, USA, pp 71110.
  • 21
    Mosher WD, Martinez GM, Chandra A, Abma JC, Wilson SJ. 2004. Use of contraception and use of family planning services in the United States: 1982–2002. No. 350. Center for Disease Control and Prevention, Hyattsville, MD, USA.
  • 22
    Johnson A, Williams R. 2004. A model to estimate influent and effluent concentrations of estradiol, estrone, and ethinylestradiol at sewage treatment works. Environ Sci Technol 38: 36493658.
  • 23
    Andersen H, Siegrist H, Halling-Sorensen B, Ternes TA. 2003. Fate of estrogens in a municipal sewage treatment plant. Environ Sci Technol 37: 40214026.
  • 24
    Baronti C, Curini R, D'Ascenzo G, Di Corcia A, Gentili A, Samperi R. 2000. Monitoring natural and synthetic estrogens at activated sludge sewage treatment plants and in a receiving river water. Environ Sci Technol 34: 50595066.
  • 25
    Johnson A, Belfroid AC, Di Corcia A. 2000. Estimating steroid estrogen inputs into activated sludge treatment works and observations on their removal from the effluent. Sci Total Environ 256: 163173.
  • 26
    Joss A, Andersen H, Ternes TA, Richie PR, Siegrist H. 2004. Removal of estrogens in municipal wastewater treatment under aerobic and anaerobic conditions. Environ Sci Technol 38: 30473055.
  • 27
    Stumpf M, Ternes TA, Haberer K, Baumann W. 1996. Nachweis von natürlichen und synthetischen Östrogenen in Kläranlagen und Flieβgewässern. Vom Wasser 87: 251261.
  • 28
    Ternes TA, Stumpf M, Muller J, Haberer K, Wilkin RD, Servos M. 1999. Behaviour and occurrence of oestrogens in municipal sewage treatment plants—I. Investigations in Germany, Canada and Brazil. Sci Total Environ 225: 8190.
  • 29
    Zuehlke S, Duennbier U, Lesjean B, Gnirss R, Buisson H. 2007. Zuehike long-term comparison of trace organics removal performances between conventional and membrane activated sludge processes. Water Environ Res 78: 24802486.

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. SUPPORTING INFORMATION
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information
FilenameFormatSizeDescription
10.1897_08-622.S1.pdf127KSupplementary Materials

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.