Burlingtonlaboratories.info

ALCOHOLISM: CLINICAL AND EXPERIMENTAL RESEARCH Ethyl Glucuronide Discloses Recent Covert Alcohol Use
Not Detected by Standard Testing in Forensic
Psychiatric Inpatients
Friedrich Martin Wurst, Rüdiger Vogel, Katja Jachau, Arthur Varga, Christer Alling, Andreas Alt, and Gregory E. Skipper Background: Considerable lives and money could be saved if one could detect early stages of lapsing/
relapsing behavior in addicted persons (e.g., in safety-sensitive workplaces) and could disclose harmfuldrinking in social drinkers. Due to the serious public health problem of alcohol use and abuse worldwide,markers of alcohol use have been sought. Both ethyl glucuronide (EtG) and phosphatidyl ethanol (PEth)appear to have high sensitivity and specificity and a time frame of detection that may elucidate alcohol usenot detected by standard testing. Our aim was to assess their potential for detecting recent covert alcoholuse under controlled conditions.
Methods: Thirty-five forensic psychiatric inpatients in a closed ward who had committed a substance-
related offense (§64 StGB), were followed for 12 months. The complete time spectrum of possible alcoholconsumption was covered by the complementary use of breath and urinary ethanol (hours), urinary EtG(days), %carbohydrate-deficient transferrin (CDT)/PEth (weeks), and ␥-glutamyltranspeptidase (GGT)/ mean corpuscular volume (MCV) (weeks-months).
Results: Fourteen of the 146 urine samples examined were positive for EtG. In all EtG-positive cases,
patients reported alcohol consumption of between 40 and 200 g of ethanol 12–60 hr prior to testing. Urinaryand breath ethanol were positive in only one case. In the blood samples, PEth was not positive in any caseand %CDT did not exceed the reference value. Isoelectric focusing showed no abnormal Tf subtypes.
Conclusions: The findings emphasize the diagnostic and therapeutic usefulness, specificity, and sensi-
tivity of EtG as a marker of recent alcohol use. Such a test is needed in numerous settings, including alcoholand drug treatment (to detect lapse/relapse), in safety-sensitive work settings where use is dangerous or inother settings where use may be inappropriate (e.g., such as driving, workplace, pregnancy, or monitoringphysicians or other professionals who are in recovery and working), or for testing other groups (such aschildren or those with medical problems) where alcohol use would be unhealthy or unsafe. The health,social and socioeconomic benefits arising from the future use of these markers is hard to overestimate.
Key Words: Biological Markers, Alcohol Drinking, Glucuronates, Ethyl Glucuronide, Physician Health
Programs, Therapy Effectiveness, Cost Reduction.
RECENTLY THE TOPIC of physicians and addiction of federally mandated workplace testing in the late 70s.
has been raised (Verghese, 2002) following the out- Testing is performed for forensic, occupational, public break of bloodstream infections in an intensive care unit. health, and therapeutic reasons. For example, State Physi- This brings to discussion the question of useful tests for cian Health Programs routinely monitor recovering physi- drug and alcohol use and monitoring individuals in safety- cians (who make a commitment to abstinence) to document sensitive jobs. Alcohol and drug testing have become com- their sobriety while the physicians continue to practice monplace in the United States following the establishment medicine.
Regulatory licensing boards have generally supported From the Psychiatric University Hospital (FMW), University of Basel, the existence of these programs, as it is known that the Basel, Switzerland; Department of Psychiatry II (FMW, RV) and Department disciplinary process is time-consuming, expensive, and as- of Legal Medicine (AA), University of Ulm, Ulm, Germany; Department of sociated with a high rate of suicide among disciplined Legal Medicine (KJ), University of Magdeburg, Magdeburg, Germany; De- licensees (Crawshaw, 1980) and, on the other hand, the partment of Medical Neurochemistry (AV, CA), University of Lund, Lund,Sweden; and the Alabama Physician Health Program and Alabama Veteri- success rates of these programs have been high (Shore, nary Professionals Wellness Program (GES), Montgomery, Alabama. 1987). Thus, urine toxicology has become an important Received for publication July 24, 2002; accepted November 14, 2002. component of these health programs. Alcohol, which re- Reprint requests: Friedrich Martin Wurst, MD, Psychiatric University Hos- mains the most common drug of choice among chemically pital, University of Basel, Wilhelm Klein Strasse 27, CH 4025 Basel, Switzer-land; Fax: 41-61-325-5583; E-mail: [email protected]. dependent health professionals (44% in Alabama), is the Copyright 2003 by the Research Society on Alcoholism. least amenable to detection. Furthermore, false-positive DOI: 10.1097/01.ALC.0000057942.57330.E2
results are possible due to fermentation in the urine post Alcohol Clin Exp Res, Vol 27, No 3, 2003: pp 471–476 voiding (Saady et al., 1993). A better means of detecting al., 2001; Wurst et al., 1995,1999a,b,2000,2002; Wurst alcohol lapse and relapse is needed.
Major shortcomings restrict the routine use of currently So far, in more than 2500 serum and urine samples available and emerging biological state markers of alcohol determined by different groups, no false positives or false consumption. These shortcomings include: (1) the time negatives have been reported (Dahl et al., 2002; Nishikawa spectrum of detection they reflect (e.g., only hours for et al., 1999; Schmitt et al., 1995, 1997; Seidl et al., 2001; serum ethanol); (2) the amount of alcohol to be consumed Wurst et al., 1999a,b,2000,2002a; Wurst and Metzger, before they are elevated (e.g., Ͼ1000 g of ethanol within 2 2002b).
weeks for carbohydrate-deficient transferrin (CDT) (Lesch Recent studies have suggested and given support to the et al., 1996; Stibler et al., 1986); (3) the availability and use of PEth in blood as a marker of alcohol abuse. Chronic practicability of the test (e.g., special labs being required alcoholics admitted for detoxification had mean PEth levels for hydroxytryptophol/hydroxyindole acetic acid (HTOL/ of 13.2 ␮mol/liter on the first day, detectable up to 14 days HIAA ratio), dolichol, acetaldehyde adducts, and so on; after admission (Hansson et al., 1997). Using liquid chro- and (4) influences on test results by other factors including matography (HPLC-ELSD) and electrospray mass spectro- age, gender, and a variety of substances and nonalcohol- metric detection, PEth has been detected in extracts of associated diseases (Gilg and Soyka, 1997; Laposata, 1999; blood from alcoholics (Gunnarsson et al., 1998). These Salaspuro, 2000). A biological state marker for disclosing patients had PEth levels of 5–13 ␮mol/liter, detectable up recent alcohol consumption with high sensitivity and spec- to 3 weeks after the beginning of an alcohol-free period. A ificity with a subacute time of detection between that of the third study on chronic alcoholics showed mean PEth levels short-term markers (such as ethanol in serum and urine, of 2.5 and 5.1 ␮mol/liter in two different groups, respec- methanol, and HTOL/HIAA ratio) and long-term markers tively (Varga et al., 2000). A study on healthy volunteers (such as CDT, GGT, or MCV) is needed.
revealed that a single dose of ethanol (32–47 g) does not Mainly in the last decade, some nonoxidative ethanol produce measurable amounts of PEth (Varga et al., 1998).
metabolites have been studied for that purpose. Promising However, out of twelve volunteers who consumed between markers include fatty acid ethyl esters (FAEE), ethyl gluc- 624 and 2134 g of ethanol during 3 weeks, eight persons had uronide (EtG), and phosphatidyl ethanol (PEth), each hav- detectable levels of PEth (1.0–2.1 ␮mol/liter). A threshold ing a characteristic time spectrum of detection of ethanol of total ethanol intake yielding detectable PEth seems to be consumption: FAEEs up to 24 hr, EtG up to 5 days, PEth around 1000 g, with a mean daily intake of about 50 g. So far, analysis of PEth has been performed by the use of up to 2 weeks (Alling et al., 1983,1984; Gunnarsson et al., whole blood. A recent study on blood from chronic alco- 1998; Hansson et al., 1997; Varga et al., 1998,2000). Be- holics showed that almost all PEth was found in the eryth- cause FAEEs, although promising (Diczfalusy et al., rocyte fraction (Varga et al., 2000).
1999,2001; Laposata and Lange, 1986), cover a time frame The aim of the present study was to further elucidate the that is included in the time spectrum of detection for EtG, potential of EtG and PEth as alcohol intake markers. This FAEEs were not determined in this study.
study was performed in the well-controlled and defined EtG is a direct metabolite of ethanol that can be conditions of a closed ward for forensic psychiatric inpa- determined in various body fluids, tissues, and hair and is tients who had committed a substance-related offense. Our nonvolatile, water soluble, and stable upon storage (Alt hypothesis was that EtG should give the best information et al., 2000; Jaakonmaki et al., 1967; Kamil et al., 1952; on recent alcohol intake and PEth, in addition to Kozu, 1973; Neubauer, 1901; Nishikawa et al., 1999; ␥-glutamyltranspeptidase (GGT), mean corpuscular vol- Schmitt et al., 1995,1997; Seidl et al., 2001; Wurst et al., ume (MCV), and %CDT could help exclude regular con- 1995,1999a,b,2000,2002; Wurst and Metzger, 2002). The sumption of greater amounts of alcohol. If EtG would turn conjugation of ethanol with activated glucuronic acid in out to be useful, this would have tremendous impact on the presence of membrane-bound mitochondrial UDP future testing for alcohol use in various settings and lead to glucuronyl transferase to form EtG represents significant improvement in therapy effectiveness, quality of 0.02– 0.06% in humans (Dahl et al., 2002) and – dose life of the patients, and socioeconomic benefit.
dependent – about 0.5–1.5% of total ethanol elimination in rabbits (Kamil et al., 1952) and, thus, a minor detox- ifying pathway. With its specific time frame of detection intermediate between short- and long-term markers and The forensic psychiatric inpatients who were studied (3 female, 32 preliminary evidence of a high sensitivity and specificity, male; median age, 33 years; range, 26–56) were all sentenced according to EtG is a promising marker of alcohol consumption. It §64 StGB (penal code), an option of German law. This law is applicable can be detected in urine beginning a few hours after for substance-related offenses when it is deemed that the individual was alcohol consumption and remains positive for up to 4 incapable, because of impairment (§20/21 StGB), of recognizing the wrongfulness of the crime. During the consecutive 12-month period of the days after the complete elimination of alcohol from the study, all patients were hospitalized on a specialized ward exclusively for body (Dahl et al., 2002; Schmitt et al., 1995,1997; Seidl et these patients. Due to the inclusion criteria (start at a given date and end Table 1. Descriptive Statistics for the Patients and Parameters Determined
PEth, phosphatidyl ethanol; EtG, ethyl glucuronide; 1, t1 ϭ month 4; 2, t2 ϭ month 11.
of study 12 months later), the patients were hospitalized between 4 days Stuttgart, Germany) as an internal standard as described elsewhere (Alt et and the entire observation period. Five were alcoholics, 17 were drug al., 1997, Wurst et al., 1999a,b,2000). The limit of determination for EtG dependent, and 13 had both diagnoses. Comorbid disorders were mainly was 0.1 mg/liter. The limit of detection was 0.03 mg/liter. Reference value: personality disorders, depressive disorders, sleep disorders, and one case 3) The determination of urinary alcohol concentration was performed on A total of 146 urine samples from 35 patients were collected randomly.
a Perkin Elmer Sigma 2000/HS 100 head-space-gas chromatograph system One to eight urine tests were performed per patient for EtG. Urine with FI detection (Perkin Elmer Sciex, Wellesley, MA).
samples were all tested for specific gravity, urinary ethanol, EtG, and illicit 4) Testing for illicit drugs was initially performed using an immunologic drugs (immunologic bedside test and REMEDI, see below). Breath eth- rapid testing device (Triage 8 test; Biosite Diagnostics, San Diego, CA) anol tests were performed randomly. In month 4 and 11 and in cases of and then the urine samples were tested at a laboratory of the Depart- suspicion of consumption blood samples were also drawn and %CDT, ment of Legal Medicine using REMEDI (Global Medical Instrumen- PEth, MCV, GGT, aspartate aminotransferase (ASAT), and serum ala- tation, Inc., Albertville, MN). The urine screening for drugs was per- nine aminotransferase (ALAT) were determined. This resulted in 43 formed by the automated HPLC-System REMEDI. The active agents samples for CDT from 26 patients (including 7 samples from one patient, were identified by online comparison of UV-spectra with a spectra data 2 samples from 11 each, and one sample from 14 individual patients). For PEth analysis, 42 samples from 26 patients were obtained (including four 5) For the determination of breath ethanol concentration, a Draeger samples from 1 patient, three samples from 2 patients, two samples from Alcotest 7410 (Draeger Safety Inc., Durango, CO) was used.
8 patients, and one sample from 16 patients). Self-reports for alcoholconsumption were obtained routinely at least once a week by interview.
1) PEth is measured in heparinized whole blood (as described elsewhere) with a high-pressure liquid chromatography (HPLC) combined with an evaporative light-scattering detector (ELSD) method (Varga et al., 2000).
All urine samples were collected under direct observation.
2) CDT estimation was performed on duplicate serum samples using the%CDT turbidimetric immunoassay kit (Bio-Rad Laboratories, Philadel- 1) To further help exclude manipulation of the urine, specific gravity and phia, PA) according to manufacturer‘s instructions. This method is based pH were determined with Combur 9 Test (Roche Diagnostics Inc., Basel, on micro anion exchange chromatography followed by turbidimetric mea- surement. Isoelectric focusing to exclude rare genetic D-variants was 2) EtG was determined with a gas chromatography-mass spectrometry undertaken as previously described followed by semiquantitative evalua- (GC/MS) method with deuterium-labeled EtG (d5-EtG) (Medichem Inc., tion by means of a scanner (Bean and James, 1994; Kuchheuser et al., 1995).
Table 2. Self-Reported Alcohol Consumption, Time of Last Drink, and Other Biomarkers Positive for Those With a Positive Urinary EtG
Breath ethanol 0.05 mg/LUrinary ethanol 0.4 g/L sumption. ROC curve analysis for EtG as test variable and self-reported alcohol consumption (yes/no) as state vari- able shows an area under the curve (AUC) of 0.964 (Fig.
No blood sample was PEth positive in any case nor did %CDT exceed the reference value (6.5). However, in four cases of those who were tested twice for %CDT, there was an increase (not exceeding the reference value) (Table 3); in four cases a decrease; and in four cases it was identical.
The Tf subtypes showed no allelic D-variants. In both samples of a patient, isoelectric focusing showed rare ge- netic Tf-subtype C2D, which generates no higher CDT.
The subtypes were: C1, 28 samples; C1–2, 10 samples; C1–3, 2 samples; C2, 1 sample; and C2D, 2 samples.
For traditional biological state markers, a good correla- tion was found between month 4 (t1) and month 11 (t2) (GGT: r ϭ 0.756, p Ͻ 0.001; MCV: r ϭ 0.895, p Ͻ 0.001).
In the Wilcoxon test, no significant differences were found for these parameters and CDT at t1 and t2.
To prevent and control health and social problems re- lated to alcohol use for the individual and society, biolog- Fig. 1. ROC curve analysis for EtG as test variable and self-reported alcohol
consumption (yes/no) as state variable. AUC ϭ 0.964.
ical state markers and marker combinations capable of monitoring alcohol consumption with a high sensitivity and III. Statistical analysis. For statistical analysis (descriptive statistics, specificity are needed. Such markers may help to (a) iden- Spearman correlation, Wilcoxon test, ROC curve analyses), SPSS 10.07 tify high risk groups; (b) evaluate current treatment pro- grams; (c) develop more effective treatment strategies; (d) disclose recent drinking in social drinkers in inappropriate and risky situations; and (e) elucidate the role of neuropsy- chological impairment following alcohol intake (i.e., the Descriptive statistics for the parameters determined (n, hangover state).
mean, median, SD, minimum, maximum) are given in Ta- The well-defined setting of a closed forensic psychiatric ble 1. During the entire study period, breath ethanol con- ward, in combination with an extensive marker battery, centration was positive in only one case at 0.05 g/liter. offered the unique opportunity to study the usefulness of Urinary ethanol was positive in the same individual at 0.4 EtG, PEth, and their combination, with traditional markers mg/liter. Specific gravity of urine was, in all cases, 1.015 and clinical impression to monitor putative alcohol con- (reference value Ͼ1.020) or higher. One urine sample was sumption in persons committed to abstinence. All patients positive for cocaine, 16 were positive for prescribed drugs studied are substance dependent and had committed a such as amitriptyline, doxepin, paroxetine, promethazine, substance-related offense. Under German law, they were trazodone, oxymetazoline, and bisoprolol. Fourteen of the committed for a term of monitoring and treatment in lieu 146 urine samples were positive for EtG (46.9 mg/dL max). of jail. During later stages of their therapy, patients were In all cases, patients reported alcohol consumption of be- allowed to leave the ward (during weekdays and on week- tween 40 and 200 g of ethanol 12–60 hr prior to testing end passes) to begin social reintegration. Despite commit- (Table 2). None of patients, from whom the 132 EtG- ment to abstinence, it was during these periods that lapses negative samples were drawn self-reported alcohol con- occurred. EtG was by far the most effective in detecting Table 3. Comparison of Different Parameters of the Four Patients With an Increase of %CDT During the Study Period
LOD, limit of determination; 1, t1 ϭ month 4; 2, t2 ϭ month 11.
surreptitious alcohol use. In contrast, traditional markers health, social, and socioeconomic benefit arising from the and clinical impression gave no indication for alcohol con- use of these markers is hard to overestimate.
sumption. Neither PEth nor %CDT was positive in any case. This supports the hypothesis that the patients were lapsing, not relapsing, and that relatively minor amounts of alcohol were being consumed. Theoretically the entire time Alling C, Gustavsson L, Änggård E (1983) An abnormal phospholipid in spectrum of possible alcohol consumption was covered by rat organs after ethanol. FEBS Lett 152:24–28.
the complementary use of ethanol (hours), EtG (days), Alling C, Gustavsson L, Månsson J, Benthi G, Änggård E (1984) Phosphatidyl-ethanol formation in rat organs after ethanol treatment.
%CDT/PEth (weeks), GGT/MCV (weeks-months). CDT and PEth are indeed similar in regarding the amount of Alt A, Wurst FM, Seidl S (1997) Bestimmung von Ethylglucuronid in alcohol that needs to be consumed for the marker to be- Urinproben mit dem internen Standard d5-Ethylglucuronid. Blutalko- come positive. The current knowledge on PEth, however, indicates that sensitivity and specificity seem to be higher. Alt A, Janda I, Seidl S, Wurst FM (2000) Determination of ethyl gluc- uronide in hair samples. Alcohol Alcohol 35:313–314.
Pharmacological interaction with prescribed or illicit Bean P, James BP (1994) Allelic D variants of transferrin in evaluation of drugs can be excluded, as all samples positive for EtG were alcohol abuse: differential diagnosis by isoelectric focusing- found to be negative for any other kind of drug. To date, immunoblotting-laser densitometry. Clin Chem 40:2078–2083.
including this study, no measurable EtG concentrations Crawshaw R (1980) An epidemic of suicide among physicians on proba- were observed in serum or urine of nonrelapsing patients, Dahl H, Stephanson N, Beck O, Helander A (2002) Comparison of nondrinking drivers, or teetotalers with negative breath urinary excretion characteristics of ethanol and ethyl glucuronide.
ethanol and/or self-reported abstinence (Nishikawa et al., 1999; Schmitt et al., 1995, 1997; Seidl et al., 2001; Wurst et Diczfalusy MA, Bjorkhem I, Einarsson C, Alexson SEH (1999) Formation al., 1999a,2000,2002; Wurst and Metzger, 2002).
of fatty acid ethyl esters in rat liver microsomes. Evidence for a key role Recent findings from the WHO/ISBRA Study on State of acyl-CoA: ethanol O-acyltransferase. Eur J Biochem 259:404–411.
Diczfalusy MA, von Wachenfeldt M, Holmberg I, Alexson SEH (2001) and Trait Markers of Alcohol Use and Dependence (WHO Possible role of long chain fatty acid ethyl esters in organ injury and as Global Status Report On Alcohol (1999); Wurst and short-term markers of ethanol intake in humans, in New and Upcoming Metzger, 2002) showed the following significant (p Ͻ 0.001) Markers of Alcohol Consumption (Wurst FM ed), pp 17–34, Steinkopff- correlations for the Spearman rank correlation for the total sample (n ϭ 304) between EtG and other variables, includ- Gilg T, Soyka M (1997) Wertigkeit biologischer Marker für Alkoholabu- sus und Alkoholismus. Nervenheilkunde 16:362–371.
ing: sobriety in days (r ϭ Ϫ0.6); the HTOL/HIAA ratio (r Gunnarsson T, Karlsson A, Hansson P, Johnson G, Alling C, Odham G ϭ 0.58); ethanol level (r ϭ 0.433); methanol level (r ϭ (1998) Determination of phosphatidyl ethanol in blood from alcoholic 0.198); CDT (r ϭ 0.458); GGT (r ϭ 0.428); and total grams males using high-performance liquid chromatography and evaporative of ethanol consumed in the previous month (r ϭ 0.467, p Ͻ light scattering of electrospray mass spectrometric detection. J Chro- 0.001). When comparing results between EtG levels (Ͼlim- matogr B Biomed Sci Appl 705:243–249.
it of determination) and the HTOL/HIAA ratio (Ͼ15 Hansson P, Caron M, Johnson G, Gustavsson L, Alling C (1997) Blood phosphatidyl ethanol of alcoholic abuse: levels in alcoholic males during pmol/nmol), 68.8% (n ϭ 119) of those positive for EtG did withdrawal. Alcohol Clin Exp Res 21:108–110.
not have elevated values for the HTOL/HIAA ratio. Thirty- Jaakonmaki PI, Knox KL, Horning EC, Horning MG (1967) The charac- one percent (31.2%) were positive for both parameters terization by gas-liquid chromatography of ethyl-␤-D-glucosiduronic- (Wurst and Metzger, 2002). This might reflect the longer acid as a metabolite of ethanol in rat and man. Eur J Pharmacol time specturm of detection for EtG (up to 5 days) as Kamil IA, Smith JN, Williams RT (1952) A new aspect of ethanol me- compared to HTOL/HIAA ratio (Ͻ1 day). These results tabolism: isolation of ethyl-glucuronide. Biochem J 51:32–33.
underline (in a big sample) the potential of EtG.
Kozu T (1973) Gas chromatographic analysis of ethyl-␤-D-glucuronide in An algorithm as suggested by the data on ethanol, EtG, human urine. Shinzu Igaku Zasshi (Med J Kagoshima Univ) 21:595– PEth, and GGT for the assessment of alcohol intake could be: (1) ethanol: recent consumption of alcohol within the Kuchheuser W, Krause D, Römhild W, Bartels H (1995) Comparative investigations on the significance of several alcoholism parameters in last day; (2) EtG: consumption not longer ago than 5 days, blood and serum samples. Rechtsmedizin 5:69–72.
detection even of small amounts like a bottle of beer; (3) Lesch O M, Walter H, Antal J, Heggli DE, Kovacz A, Leitner A, Neumeis- PEth: 1000 g of ethanol to be consumed within 2 weeks; ter A, Stumpf I, Sundrehagen E, Kasper S (1996) Carbohydrate- and (4) GGT: consumption of greater amounts over a deficient transferrin as a marker of alcohol intake: a study with healthy subjects. Alcohol Alcohol 31:265–271.
Laposata M (1999) Assessment of ethanol intake–current tests and new A test like that for EtG is needed in numerous settings, assays on the horizon. Am J Clin Pathol 112:443–450.
including alcohol and drug treatment (to detect lapse/re- Laposata EA, Lange LG (1986) Presence of nonoxidative ethanol metab- lapse), in safety-sensitive work settings where use is dan- olism in human organs commonly damaged by ethanol abuse. Science gerous or in other settings where use may be inappropriate (e.g., such as driving, workplace, pregnancy), or for testing Neubauer O (1901) Ueber Glucuronsäurepaarung bei Stoffen der Fett- reihe. Arch Exp Path Pharmak 46:133–154.
other groups (such as children or those with medical prob- Nishikawa M, Tsuchihashi H, Miki A, Katagi M, Schmitt G, Zimmer H, lems) where alcohol use would be unhealthy or unsafe. The Keller TH, Aderjan R (1999) Determination of ethyl glucuronide, a minor metabolite of ethanol, in human serum by liquid chromatography- Varga A, Hansson P, Johnson G, Alling C (2000) Normalization rate and electrospray ionization mass spectrometry. J Chromatogr B Biomed Sci cellular localization of phosphatidyl ethanol in whole blood from chronic alcoholics. Clin Chim Acta 299:141–150.
Saady JJ, Poklis A, Dalton HP (1993) Production of urinary ethanol after Verghese A (2002) Physicians and addiction. N Engl J Med 346:1510– sample collection. J Forensic Sci 38:1467–1471.
Salaspuro M (2000) Marker für Alkoholismus und Alkoholassoziierte WHO Global Status Report On Alcohol (WHO/HSC/SAB/99.11) (1999) http://www.who.int/substance_abuse/pubs_alcohol.htm (accessed Sep- Organschäden, in Handbuch Alkohol, Alkoholismus, Alkoholbedingte Organschäden (Seitz HK, Lieber CS, Simanowski UA eds), pp 96–125, Wurst FM, Seidl S, Sachs H, Besserer K, Hirsch M, Reinhardt G, Schüt- Johann Ambrosius Barth, Leipzig-Heidelberg.
tler R (1995) Ethyl glucuronide–evaluation of a new marker of alco- Schmitt G, Aderjan R, Keller Th, Wu M (1995) Ethyl-Glucuronide: an holism (abstract). Alcohol Alcohol 30:560.
unusual ethanol metabolite in humans. Synthesis, analytical data, and Wurst FM, Kempter CH, Seidl S, Alt A (1999a) Ethyl glucuronide–a determination in serum and urine. J Anal Toxicol 19:91–94.
marker of alcohol consumption and a relapse marker with clinical and Schmitt G, Droenner P, Skopp G, Aderjan R (1997) Ethyl glucuronide forensic implications. Alcohol Alcohol 34:71–77.
concentrations in serum of human volunteers, teetotalers, and sus- Wurst FM, Schüttler R, Kempter CH, Seidl S, Gilg TH, Jachau K, Alt A pected drinking drivers. J Forensic Sci 42:1099–1102.
(1999b) Can ethyl glucuronide be determined in post-mortem body Seidl S, Wurst FM, Alt A (2001) Ethyl glucuronide–a biological marker fluids and tissues? Alcohol Alcohol 34:260–261.
for recent alcohol consumption. Addiction Biol 6:205–212.
Wurst FM, Kempter CH, Metzger J, Seidl S, Alt A (2000) Ethyl glucu- Shore JH (1987) The Oregon experience with impaired physicians on ronide–a marker of recent alcohol consumption with clinical and foren-sic implications. Alcohol 20:111–116.
Wurst FM, Metzger JW on behalf of the WHO/ISBRA Study on State and Stibler H, Borg S, Joustra M (1986) Micro anion exchange chromatography Trait Markers of Alcohol Use and Dependence Investigators. (2002) of carbohydrate-deficient transferrin in serum in relation to alcohol con- The ethanol conjugate ethyl glucuronide is a useful marker of recent sumption (Swedish Patent 8400587–5). Alcohol Clin Exp Res 10:535–544.
alcohol consumption. Alcohol Clin Exp Res 26:1114–1119.
Varga A, Hansson P, Lundqvist C, Alling C (1998) Phosphatidyl ethanol Wurst FM, Seidl S, Ladewig D, Müller-Spahn F, Alt A (2002) Ethyl in blood as a marker of ethanol consumption in healthy volunteers: glucuronide on the time course of excretion in urine during detoxifica- comparison with other markers. Alcohol Clin Exp Res 22:1832–1837.

Source: http://burlingtonlaboratories.info/assets/Uploads/EtG.ForensicPsychatrics.pdf