Anal Bioanal Chem DOI 10.

1007/s00216-008-2546-2

ORIGINAL PAPER

Microwave-assisted extraction: a simpler and faster method for the determination of ethyl glucuronide in hair by gas chromatographymass spectrometry
Ivn lvarez & Ana Mara Bermejo & Mara Jess Tabernero & Purificacin Fernndez & Pamela Cabarcos & Patricia Lpez

Received: 2 October 2008 / Revised: 6 November 2008 / Accepted: 25 November 2008 # Springer-Verlag 2008

Abstract Alcohol is the most frequently abused addictive substance that causes serious social problems throughout the world; thus, alcoholism is of particular interest in clinical and forensic medicine. Alcohol biomarkers are physiological indicators of alcohol exposure or ingestion and may reflect the presence of an alcohol use disorder. The glucuronide conjugation is a minor pathway of ethanol metabolism. Ethyl glucuronide (EtG) is a marker of recent alcohol consumption that detects alcohol use reliably over a definite time period. The present paper describes a new method for the determination of EtG in hair. It is based both in the microwave-assisted extraction (MAE), to extract the analyte from hair samples, and gas chromatographymass spectrometry (GC-MS), to identify and quantify the EtG in selected ion monitoring (SIM) mode. The method was applied to 15 hair samples from occasional alcohol users, obtaining positive results in all cases. It was fully validated, including a linear range (0.310 ng/mg) and the main precision parameters. In summary, the use of microwave-assisted extraction turned out to be a substantially simpler, faster, and a more sensitive procedure than any other conventional sample preparations. Keywords Ethyl glucuronide . Hair . Microwave-assisted extraction . Gas chromatographymass spectrometry

Introduction Alcohol is a depressant drug that has been produced and drunk in Europe for thousands of years [1]. It has traditionally been considered a serious public health problem because of several physical, mental and social problems associated with its abuse, as it exerts an enormous toll on the lives and communities of many nations [2]. In fact, the Spanish 2005 National Report, presented to the European Monitoring Centre for Drugs and Drug Addiction, has shown that alcohol is by far the psychoactive substance of abuse most widely consumed by the Spanish population [3]. Detection of ethanol in body fluids is only possible during a relatively short time after alcohol consumption. About 9095% of alcohol is eliminated by oxidation mainly in the liver, whereas biotransformation of ethanol to ethyl glucuronide (ethyl--D-6 glucuronic acid; EtG), via conjugation with activated glucuronic acid, represents only 0.020.06% of complete alcohol elimination [4, 5]. Although the percentage of alcohol metabolized by this later pathway is small, it represents a useful tool as EtG becomes detectable up to 4 days after complete elimination of alcohol from the body. With its specific time frame of detection, intermediate between short- and long-term markers, and a high sensitivity and specificity (i.e., EtG is not detectable unless alcohol has been consumed), EtG is a promising marker of alcohol consumption in general and a marker for relapse detection [6, 7]. Its detection in hair is more and more studied during the last years for the purpose of alcohol abuse monitoring, in both clinical and forensic toxicology [8]. It was determined by GC-MS-EI, by GC-MS/NCI, and by LC-MS/MS [4]. Generally, some extraction methods have been used in forensic toxicology, such as liquidliquid extraction, solid phase extraction or solid phase microextraction and so on

I. lvarez : A. M. Bermejo (*) : M. J. Tabernero : P. Fernndez : P. Cabarcos : P. Lpez Institute of Legal Medicine, Forensic Toxicology Service, Faculty of Medicine, C/ San Francisco s/n, 15782 Santiago de Compostela, Spain e-mail: anamaria.bermejo@usc.es

I. lvarez et al.

[9, 10, 11]. But some need lots of time, some are expensive and some require large amounts of organic solvents. Simple, rapid and less labor-intensive extraction techniques are needed in forensic toxicology. The growing interest in obtaining increasingly better results in this context has led to the development of microwave-assisted extraction (MAE), which has been widely used for the extraction of organic pollutants from sediments, soil, water, and other types of materials. By contrast, MAE has scarcely been used to extract drugs of abuse from different materials such as serum, urine, tablets, or coca leaves [12, 13]. Commercial microwave equipments with security systems and closed vessels have made MAE an analytical technique of interest since these instruments enable simultaneous extractions of analytes at high pressure and temperature. The MAE technique requires less solvent consumption and shorter extraction times, while the extraction yields of the analyte are equivalent to or even higher than those obtained with conventional methods. The MAE has been applied successfully to the extraction of ethyl glucuronide from urine by our investigative group, and this method has been incorporated to our laboratory routine [14]. Because of these issues, this work proposes a new, fast, sensitive, and reliable method to detect ethyl glucuronide in human hair samples by using MAE and GC-MS.

samples simultaneously in PTFE -lined extraction closed vessels under the same conditions (temperature and pressure), with simultaneous magnetic stirring of the sample and solvent inside. An in-board control system was installed for monitoring and controlling pressure and conditions inside the extraction vessels. This oven allows a maximum of 1,000 W and the power changes in order to reach and maintain the temperature selected. Chromatographic analyses for EtG were performed using an electron impact ionization gas chromatograph model 6890 from Hewlett-Packard (Little Falls, DF, USA) interfaced to a mass selective detector (MSD) model 5973 inert from Agilent Technologies (Las Rozas, Spain). Analytical conditions Chromatographic elution was performed using the previously published fully validated method described above [14]. Initially neat standards of EtG and EtG-d5 (2 l of a 0.01 mg/mL solution) were injected in a mixture after derivatization and analyzed using the full scan mode of the GC-MS. Quantifier and qualifier ions used for each analyte were selected based on their abundance and m/z values. Because of their reproducibility and lack of interference, high mass ions were selected when possible. The ions selected were as follows: m/z 160, 261, 405 (EtG); m/z 165, 266, 410 (EtG-d5), using the underlined ones for quantitation. Upon selection of unique ions, the MS was run in selected ion monitoring (SIM) mode. Extraction procedure and derivatization

Experimental Materials and methods Chemical reagents and standards Ethyl glucuronide and ethyl glucuronide-d5 were purchased from Medichem (Stuttgart, Germany). Methanol gradient grade, chloroform, pyridine and N,O-(bistrimethylsilyl) trifluoroacetamide (BSTFA) were obtained from Merck (Darmstadt, Germany). Preparation of solutions Stock standard solutions of EtG and EtG-d5 at a concentration of 1 mg/mL in methanol were prepared, and from these, working solutions were obtained by dilution with methanol. Samples were stored at 4 C when not in use. Distilled water was processed through a Milli-Q water system (Millipore, Bedford, MA, USA). Instrumentation The microwave extractor system was an ETHOS PLUS MPR300/12S (Milestone, Agrigento, Italy) equipped with a solvent detector. The microwave was able to extract 12

Hair samples obtained from subjects who did not drink any alcohol were used to carry out the calibration curves. Hair samples were submitted to an initial procedure of decontamination by washing twice in 5 mL of a 0.1% solution of Tween 80 for 10 min, and rinsing twice with 5 mL of distilled water to eliminate external contamination, and the last wash analyzed to exclude possible interferences, e.g., of any cosmetic product. After drying, each sample was cut in 1-mm segments and then 100 mg of hair were weighted. Thirty microliters of the deuterated internal standard (10 g/ml) were added to 100 mg of hair. The internal standard was used to eliminate injection error while maintaining a constant area ratio for concentration quantitation. The sample was mixed with 8 mL of n-hexan/water (1:1 v/v) and placed in the vessel of the microwave oven for extraction at 110 C for 11 min. After the extraction, the vessel contents were centrifuged at 4,000 rpm for 5 min. The aqueous layer was removed by evaporation to dryness under a nitrogen stream in a thermostatic bath at 50 C. Derivatization was performed using the previously published method described above [14].

Microwave-assisted extraction: a simpler and faster method for the determination of ethyl glucuronide in hair Fig. 1 a Chromatogram for a blank hair sample. b Chromatogram for a hair sample spiked with EtG (1 ng/mg). c Chromatogram of a real sample (EtG: 0.53 ng/mg)

I. lvarez et al. Table 1 Limit of detection, lower limit of quantitation and calibration results of EtG LOD (ng/mg) 0.1 LLOQ (ng/mg) 0.3 Linearity y=0.2466x+0.1061 Slope standard error 0.00459 Intercept standard error 0.02450 R 0.990

Validation of the method The analytical validation of the method was performed by establishing selectivity, linearity, limits of detection and quantitation, intra- and inter-day precision and accuracy, and recovery [10, 11]. Selectivity of the method was demonstrated by analyzing ten hair samples from different people who had not consumed ethanol. Samples were extracted and analyzed for assessment of potential interferences from endogenous substances. The apparent response at the retention times of the analytes under investigation was compared with the response of analytes at the limit of quantitation. Standard calibration curves were obtained as previously described using drug-free control hair spiked with standard solution to obtain the concentration range of 0.310 ng/mg for EtG. Quantitation was based on target peak area ratio of EtG (m/z 261) to its internal standard (m/z 266). Sensitivity of the method was determined by calculation of the limit of detection (LOD) and the lower limit of quantitation (LLOQ). LOD was determined by an empirical method that consists of analyzing a series of hair samples containing decreasing amounts of the analytes. LOD was the lowest concentration that presented a signal-to-noise ratio higher than 3 for at least three diagnostic ions for each substance. The LLOQ is the lowest concentration of analyte that can be determined quantitatively with appropriate precision and accuracy. Precision and accuracy were determined by inter- and intra-day assays. Inter-day precision and accuracy were evaluated by six determinations per concentration in differ-

ent days. Intra-day precision and accuracy were determined at three concentrations, 0.3, 3, and 10 ng/mg, by preparing and analyzing on the same day five replicates for each level. Precision, expressed as the coefficient of variation (CV) of the measured values, calculated as (standard deviation/ mean)100, is expected to be less than 15% at all concentrations, except for the LLOQ for which 20% is acceptable. In the same way, accuracy was evaluated using the mean relative error (MRE), which had to be less than 15% of the theoretical values at each concentration level except for the LLOQ, for which 20% is acceptable [15]. Recovery or extraction efficiency (%) for the analyte was determined at low and high concentration levels. Calculations were performed by comparing the areas of the peaks after extraction of samples with the internal standard and the EtG, with those obtained containing only the internal standard and subsequently spiked with the drug at the same concentration.

Results and discussion Sample preparation is one of the most important steps in the majority of analytical procedures to determine constituents in samples with complex matrices. An ideal sample preparation technique should be simple, inexpensive, efficient, selective, and compatible with various analytical techniques. It should give as high a recovery as possible, use the minimum amount of solvent, and be environmentally friendly. Microwave-assisted extraction has risen rapidly in the last decade, and for most applications it has proven to be effective in all the aspects mentioned above compared to traditional extraction procedures. It is well known that ethylglucuronide can accumulate in hair, where it can be detected following alcohol consumption. Furthermore, hair is easy to obtain through non-invasive procedures. Some studies have demonstrated that EtG is highly specific and sensitive as an alcohol marker [16].

Table 2 Precision and accuracy obtained for the EtG in hair Concentration Intra-day study (n=5) Inter-day study (n=6) added (ng/mg) C.V. (%) Relative mean C.V. (%) Relative mean error (%) error (%) 0.3 0.5 1 3 5 8 10 6.76 5.35 9.32 7.95 4.36 5.11 6.12 5.59 5.90 4.77 9.64 13.43 5.00 0.78 1.88 0.69

Table 3 Recoveries of EtG in hair (n=5) Concentration (ng/mg) 10 70 Mean Recovery (%) 86.54 88.45 CV (%) 2.87 11.86

3.81

7.71

2.17

3.59

Microwave-assisted extraction: a simpler and faster method for the determination of ethyl glucuronide in hair Table 4 Results for the 15 analyzed hair samples Case 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 EtG (ng/mg) 0.44 3.09 1.87 0.79 1.56 2.93 0.83 1.96 4.59 2.77 2.32 3.72 4.20 3.58 0.53

In the consulted literature, it has been possible to find some papers in which EtG was extracted from hair. In most of them, ultrasonication for 2 or 3 h is required and a solid phase extraction procedure of approximately 45 min is needed. Some authors even describe a previous step of incubation at 2530 C during 5 and 12 h [2, 4, 1621]. This kind of extraction is more laborious than the one described in our paper, with a higher time and solvent consumption. Silylation is the most widely used derivatization procedure for sample analysis by GC. The popularity of silylation reagents is enhanced by their ease of use and formation of derivatives. In silylation, an active hydrogen is replaced by an alkylsilyl group such as trimethylsilyl (TMS). Compared to their parent compounds, silyl derivatives are more volatile, less polar, and more thermally stable. As a result, GC separation and detection is improved. Silylation reagents are generally moisture sensitive, requiring them to be sealed to prevent deactivation. Some authors compared several derivatizing agents for EtG like PFPA (pentafluopropionic anhydride), but in our case, we selected the mixture obtained with BSTFA and pyridine, and the silyl derivative obtained showed to be stable for more than 4 h [9, 14]. BSTFA attacks the hydroxyl groups of the ethylglucuronide and the result is a volatile trimethylsilyl derivate, thus making it easily detected in the GC-MS. Figure 1a shows a chromatogram obtained for a blank hair sample, and shows no significant interferences of any peak appearing at the expected retention time for the analyte (12.80 min), demonstrating a good selectivity of the proposed method. Figure 1b shows a chromatogram for a hair sample spiked with EtG. The calibration plot was linear for the EtG over the specific range (0.310 ng/mg). A simple linear regression

analysis was performed. The LOD, LLOQ, and calibration results are detailed in Table 1. The confidence parameters of the validated method (inter- and intra-day precision and accuracy) for the determination of the studied analyte are shown in Table 2. Precision and accuracy of the analyte under investigation at reported concentrations satisfactorily met the international established acceptance criteria [15]. The recoveries obtained are presented in Table 3. These results suggest that at a low analyte concentration, the mean recoveries are slightly lower than at a high analyte concentration. Stability of the derivatized analyte was studied for a period of 4 h for the highest concentration. A constant behavior was observed throughout this period. The analytical procedure proposed for the determination of ethyl glucuronide in hair showed to be highly precise with the use of the respective deuterated internal standard. Good sensitivity and linearity were also obtained for the analyte. Finally, the developed method was used to analyze 15 real hair samples that were received in our laboratory (Forensic Toxicology Service, Institute of Legal Medicine, University of Santiago de Compostela) from people who acknowledged to be occasional alcohol users (Table 4). Results show that in all cases analyzed it was possible to detect EtG, confirming its usefulness as a marker of alcohol consumption and obtaining a wide range of concentrations (0.44 to 4.89 ng/mg). A representative chromatogram of a real sample is presented in Fig. 1c.

Conclusions The GC-MS method reported in this paper to analyze ethyl glucuronide in hair was validated in the range (0.310 ng/ mg), according to internationally acceptance criteria and it was found to be sensitive enough to quantify 0.3 ng/mg. The proposed MAE method is being routinely used for the determination of EtG in hair in our laboratory. Microwave energy expedites extraction of the analyte while maintaining its recovery rates. Because of the technique uses standard laboratory equipment, it can be an effective method relative to potential alternatives such as liquid liquid extraction and solid phase extraction. In fact, MAE reduces solvent consumption and extraction time, with respect to others published methods. MAE is an extraction procedure that will also be useful when applied to more sensitive instrumentation, such as GC/MS/MS, LC/MS, and all types of analytical techniques. GC-MS was found to be specific, sensitive, and selective enough for determining the low analyte concentrations to be expected in hair. So, the method meets the sensitivity and the selectivity requirements for clinical and forensic toxicology.

I. lvarez et al.

References
1. Anderson P, Baumberg B (2006) Alcohol in Europe. A public health perspective. A report for the European Commission. Institute of Alcohol Studies. United Kingdom 2. Politi L, Zucchella A, Morini L, Stramesi C, Polettini A (2007) Forensic Sci Int 172(1):2327 3. Spanish 2005 National Report to the European Monitoring Centre for Drugs and Drug Addiction. European Monitoring Centre for Drugs and Drug Addiction. [Available on line at URL: http:// ar2006.emcdda.europa.eu/es/home-es.html] (Accessed April 2007) 4. Yegles M, Labarthe A, Auwrter V, Hartwig S, Vater H, Wennig R, Pragst F (2004) Forensic Sci Int 145(23):167173 5. Wurst FM, Skipper GE, Weinmann W (2003) Addiction 98(Suppl. 2):5161 6. Wurst FM, Metzger J (2002) Alcohol Clin Exp Res 26(7):11141119 7. Wurst FM, Seidl S, Ladewig D, Mller-Spahn F, Alt A (2002) Addict Biol 7:427434 8. Kissack JC, Bishop J, Leatherwood Roper A (2008) Pharmacotherapy 28(6):769781 9. lvarez I, Palos F, Bermejo AM, Fernndez P, Tabernero MJ (2006) Anal Letters 39:13931399 10. lvarez I, Bermejo AM, Tabernero MJ, Fernndez P, Lpez P (2007) J Chromatogr B 845:9094

11. Lpez P, Bermejo AM, Tabernero MJ, Fernndez P, lvarez I (2007) J Appl Toxicol 27(5):464471 12. Fernndez P, Lago M, Lorenzo RA, Carro AM, Bermejo AM, Tabernero MJ (2007) J Appl Toxicol 27:373379 13. Fernndez P, Lago M, Lorenzo RA, Carro AM, Bermejo AM, Tabernero MJ (2007) J Anal Toxicol 31:388393 14. lvarez I, Bermejo AM, Cid P, Tabernero MJ, Fernndez P, Lpez P (2008) J Appl Toxicol 28:773778 15. Food and Drug Administration: U.S. Department of Health and Human Services, Guidance for Industry, Bioanalytical Method Validation. FDA [Available on line at URL: http:// www.fda.gov/cder/guidance/4252fnl.pdf] (Accessed April 2006) 16. Wurst FM, Kempter C, Metzger J, Seidl S, Alt A (2000) Alcohol 20:111116 17. Janda I, Weinmann W, Kuehnle T, Lahode M, Alt A (2002) Forensic Sci Int 128:5965 18. Skopp G, Schmitt G, Ptsch L, Drnner P, Aderjan R, Mattern R (2000) Alcohol Alcohol 35(3):283285 19. Kintz P, Villain M, Vallet E, Etter M, Salquebre G, Cirimele V (2007) Forensic Sci Int 176(1):8790 20. Paul R, Kingston R, Tsanaclis L, Berry A, Guwy A (2008) Forensic Sci Int 176(1):8286 21. Appenzeller BMR, Agirman R, Neuberg P, Yegles M, Wennig R (2007) Forensic Sci Int 173:8792