Speciation Analysis of Gadolinium Chelates in Hospital Effluents and

Mar 18, 2009 - Marvin Birka , Kristina S. Wentker , Elke Lusmöller , Brigit Arheilger , Christoph A. Wehe , Michael Sperling , Rudolf Stadler , and U...
0 downloads 0 Views 1MB Size
Environ. Sci. Technol. 2009, 43, 2884–2890

Speciation Analysis of Gadolinium Chelates in Hospital Effluents and Wastewater Treatment Plant Sewage by a Novel HILIC/ICP-MS Method ¨ NNEMEYER, LYDIA TERBORG, JENS KU ¨ RN MEERMANN, BJO CHRISTINE BRAUCKMANN, ¨ LLER, ANDY SCHEFFER,† AND INES MO UWE KARST* Westfa¨lische Wilhelms-Universita¨t Mu ¨ nster, Institut fu ¨r Anorganische and Analytische Chemie, Corrensstr. 30, 48149 Mu ¨ nster, Germany

Received November 24, 2008. Revised manuscript received February 23, 2009. Accepted February 23, 2009.

The behavior of Gd chelates used in magnetic resonance imaging (MRI) within the process of sewage treatment is widely unknown. Due to the varying toxicity of the particular Gd species [J. M. Idee et al. Fundam. Clin. Pharmacol. 2006, 20, 563-576], it is important to not only investigate total Gd concentrations, but the Gd species as well. This work describes a novel method for speciation analysis of the most important gadolinium chelates in wastewaters. This novel approach consists of coupling hydrophilic interaction chromatography (HILIC) with inductively coupled plasma mass spectrometry (ICP-MS). HILIC/ICP-MS exhibits high separation efficiency for the simultaneous separation of the five predominantly applied MRI contrast agents and the required selectivity and sensitivity for trace determination in wastewater samples. For the first time, the distribution of particular Gd chelate complexes was determined in hospital effluent, municipal sewage, and wastewater treatment plant (WWTP) samples. The data were compared with the total concentration of Gd as determined by ICP-MS. The active compounds of Multihance, Dotarem, and Gadovist were identified in local WWTP samples. Interestingly, the macrocyclic, nonionic compound Gd-BT-DO3A (Gadovist) was found to be the most abundant Gd complex in all investigated samples. This is in contrast to prevalent assumptions that linear ionic Gd chelates such as Gd-DTPA (Magnevist) would be the predominant species [G. Morteani et al. Environ. Geochem. Health 2006, 28, 257-264 and M. Bau and P. Dulski, Earth Planet. Sci. Lett. 1996, 143, 245-255]. Although contrast agent concentrations tend to be reduced during wastewater treatment, Gd-BT-DO3A was still found in WWTP effluents.

Introduction Since their introduction in the 1980s, the application of Gdbased contrast agents increased rapidly. Today, almost every * Corresponding author phone: +49-251-8333141; fax: +49-2518336013; e-mail: [email protected]. † Current address: Niedersa¨chsisches Landesamt fu ¨r Verbraucherschutz und Lebensmittelsicherheit (LAVES), Futtermittelinstitut Stade, Heckenweg 4-6, 21680 Stade, Germany. 2884

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 8, 2009

second MRI procedure is enhanced by contrast agents, resulting in about 20,000,000 applications per year (1). Gadolinium has seven unpaired 4f electrons and thus is one of the most paramagnetic metals. The large magnetic momentum of the Gd3+ ions can polarize water molecules in their vicinity, leading to accelerated spin relaxation (T1) of the water protons and to increased MRI signal intensity. As particular body tissues have different water contents, Gdbased contrast agents exhibit excellent magnetic properties to enhance the image contrast for MRI of fine body structures. For the application as contrast agent, gadolinium is usually complexed by polyaminocarboxylic acid chelating agents. The five predominantly applied Gd complexes used as contrast agents are shown in Figure 1. In general, contrast agent formulations are highly concentrated (0.5-1.0 mol/L Gd) so that on an average, approximately 1.2 g of Gd is applied to a typical MRI patient with each dose. This leads to a very high input of anthropogenic Gd into the environment. Ku ¨ mmerer et al. calculated that, regarding the total number of MRI systems used in Germany, the Gd emission by hospitals and practices exceeds 1,100 kg per year (4). The extensive accumulation of Gd in the environment was observed by Bau and Dulski in 1996 (3). They investigated rare earth element (REE) distributions in numerous water samples of different origin and found anomalously high concentrations of Gd. In general, REE metals behave coherently in natural systems. When normalized to their natural background concentrations, semilogarithmic plots of normalized concentration vs REE metal exhibit smooth patterns. However, Bau et al. found REE patterns with drastically elevated Gd concentrations for water samples from densely populated areas with highly developed medical healthcare systems (1, 5). Positive Gd anomalies have also been found in numerous rivers and surface waters from Europe, Asia, America, and Australia (2, 3, 6-14). Extraordinarily high Gd anomalies were observed in Berlin (3, 6). Moreover, positive Gd anomalies were even observed in North Sea water (13). Even though concentrations are much lower than those used in medical applications, long-term effects on the aquatic environment, biota, and humans are still unknown and potential toxic effects cannot be dismissed (8, 15, 16). Gd-based contrast agents are excreted in unmetabolized form and are suspected not to be biodegradable in aqueous test systems (4). Due to their high polarity and good water solubility they exhibit a long half-life (13). Until now, the behavior of Gd chelates within the process of sewage treatment has been widely unknown (4). It is assumed that Gd is still bound in stable aqueous complexes. However, as Fe3+ is used during the flocculation process within wastewater treatment, nontoxic complexed Gd from contrast agents may possibly be transmetallated to uncomplexed toxic species due to competitive Fe3+ cations (3, 4). To corroborate this assumption and to clarify the behavior of anthropogenic gadolinium during wastewater treatment, there is a need for analytical methods for the speciation analysis of particular Gd chelates in water samples. Classically, the analysis of REEs is performed by sequential partitioning into particulate, colloidal, and dissolved fractions, followed by preconcentration and determination by inductively coupled plasma (ICP) mass spectrometry (MS), or ICP with optical emission spectroscopy (OES) (17, 18). In contrast, trace determination of particular Gd species often requires combination of powerful separation techniques with sensitive and element-selective detectors. Kru ¨ ger et al. (19) and Loreti and Bettmer (20) used size exclusion chromatography (SEC) with ICP-MS to separate 10.1021/es803278n CCC: $40.75

 2009 American Chemical Society

Published on Web 03/18/2009

FIGURE 1. Chemical structures of the five predominantly applied contrast agent complexes and the trademarks of the respective drugs. free Gd3+ from complexed Gd and Gd adducts. However, size differences between the individual Gd chelates are too small for an efficient separation by SEC. Analysis of free REE cations, utilizing ion exchange chromatography (IEC) coupled to ICP-OES (21) and ICP-MS (22-24) was described by Yoshida, Wang, Alonso, and Kerl. Although reversed phase chromatography (RP-HPLC) exhibits only very little retention for polar compounds, Mazzucotelli et al. showed the separation of one contrast agent and a number of degradation products by RP-HPLC/ICP-MS (25). Kautenburger et al. used the high separation efficiency of capillary electrophoresis (CE) for the analysis of various Gd species complexed or uncomplexed with humic substances by CE/ICP-MS (26, 27). The separation of all lanthanides by CE/ICP-MS was reported by Day et al. (28). Hydrophilic interaction chromatography (HILIC) offers high separation efficiency for polar and ionic compounds and was successfully applied for the determination of contrast agents in blood plasma samples by HILIC with electrospray ionization mass spectrometry (ESI-MS) in an earlier publication of our group (29). However, ESI-MS does not offer the required sensitivity for Gd speciation in wastewater samples. Moreover, common eluents in HILIC consist of aqueous solutions with very high organic content, which is rather unfavorable for ICP-MS detection. Nevertheless, the coupling of HILIC with ICP-MS was first described by Ouerdane et al. (30) and Dernovics and Lobinski (31). Soon thereafter, Hemstro¨m et al. (32) and Nygren et al. (33) applied HILIC/ ICP-MS for the analysis of nickel, selenium, and platinum species. This manuscript describes the development and application of a novel HILIC/ICP-MS method for speciation analysis of the most important Gd-based MRI contrast agents in hospital effluents and WWTP sewage.

Materials and Methods Chemicals and Consumables. Diethylenetriaminepentaacetic acid (DTPA) and diethylenetriaminepentaacetic acid gadolinium(III) dihydrogen salt hydrate (Gd-DTPA) were purchased from Sigma-Aldrich Chemie GmbH (Steinheim, Germany). Formic acid, ammonium formate, and ammonium hydroxide solution (25%) were purchased from Fluka Chemie GmbH (Buchs, Switzerland). Rhodium standard (10 mg/L Rh), gadolinium standard (1000 mg/L Gd) for ICP-MS (traceable to NIST SRM 3118a, lot 992004), methanol,

and acetonitrile for HPLC were obtained from Merck KGaA (Darmstadt, Germany). Hexamethyldisilazane (99%) was obtained from ABCR (Karlsruhe, Germany). All contrast agent infusion solutions were obtained from their respective pharmaceutical companies: Gadovist (Gd-BT-DO3A, 1.0 mol/ L) and Magnevist (Gd-DTPA, 0.5 mol/L) from Bayer Schering Pharma AG (Berlin, Germany), Omniscan (Gd-DTPA-BMA, 0.5 mol/L) from GE Healthcare Buchler (Braunschweig, Germany), Dotarem (Gd-DOTA, 0.5 mol/L) from Guerbet (Sulzbach, Germany), and Multihance (Gd-BOPTA, 0.5 mol/ L) from Nycomed GmbH (Konstanz, Germany). All chemicals were used in the highest quality available. Water used for HPLC was purified using a Milli-Q Gradient A 10 system and filtered through a 0.22 µm Millipak 40 filter (Millipore, Billerica, MA). Nylon syringe filters (25 mm × 0.45 µm) were obtained from Alltech/Grace, Deerfield, IL). Glassware Pretreatment. To circumvent sorption effects of the polar analytes to the glass surface, all glassware was inactivated by treatment with hexamethyldisilazane (HMDS), employing a modified method of Fenimore et al. (34). Volumetric flasks were thoroughly cleaned with commercially available alkaline agent, rinsed with purified water, and dried in a drying cabinet prior to silylation. HPLC vials were stored in a large glass flask under HMDS atmosphere, whereas volumetric flasks were directly treated with 0.5 µL of HMDS per mL of flask volume. The glassware was sealed and stored overnight at a temperature of 80 °C. To evaporate excess HMDS, stoppers were removed and after additional 90 min at 80 °C the glassware was cooled to ambient temperature and finally rinsed with purified water to hydrolyze any unreacted HMDS. Stock and Standard Solutions. Stock solutions with a concentration of 10 mmol/L were prepared by dilution of the respective infusion solution with purified water. A 10 mM DTPA stock solution was prepared by dissolving the free acid in purified water. To improve the solubility of DTPA, 1% of ammonium hydroxide solution (25%) was added. HILIC/ ICP-MS calibration solutions were prepared by dilution of appropriate amounts of the respective 10 mM contrast agents. For HILIC/ICP-MS measurements, six calibration solutions were prepared in the range from 1 × 10-9 to 5 × 10-8 mol/L (0.157-7.863 µg/L Gd), containing Gd-BOPTA, Gd-DTPABMA, Gd-BT-DO3A, Gd-DOTA, and Gd-DTPA. To cope with the same class of analytes in ICP-MS analysis, Gd-DTPA (10 VOL. 43, NO. 8, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2885

FIGURE 2. Schematic drawing of the Mu¨nster main wastewater treatment plant (WWTP).

TABLE 1. Sample Abbreviations, the Respective Sampling Sites As Shown in Figure 2, Molar Concentrations of the Individual Gd Species, Standard Deviation of the Mean Values of the Triplicate Measurements and Sum of All Gd Species Determined by HILIC/ ICP-MS; Total Molar Concentration of Gd Determined by ICP-MS HILIC/ICP-MS Gd-BOPTA

Gd-BT-DO3A

Gd-DOTA

Σ Gd

Gdtot

sampling site

-9

-9

-9

-9

c × 10-9 [mol/L]

UKM-W01; hospital, western tower UKM-W02; hospital, western tower UKM-E03; hospital, eastern tower UKM-E04; hospital, eastern tower SEW05; sewage system WWTP06; inlet construction (a) WWTP07; inlet construction (a) WWTP08; primary clarifier (d) WWTP09; primary clarifier (d) WWTP10; dephosphatation (g) WWTP11; nitrification/denitrification (j) WWTP12; secondary clarifier (k) WWTP13; secondary clarifier (k) WWTP14; outlet analyzer (l)

c × 10

1.1 1.2

[mol/L]

((0.1) ((0.1)

c × 10

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 8, 2009

[mol/L]