by Hamid R. Tavakoli, MD, FAPA, FAPM; Michael Hull, BS; and LT. Michael Okasinski, MD
Dr. Tavakoli is Chief, Consultation-Liason Service, Naval Medical Center, Portsmouth, Virginia, and Assistant Professor, Eastern Virgina Medical School; Mr. Hull is a medical student at Eastern Virginia Medial School; and Dr. Okasinki is staff psychiatrist at US Naval Hospital, Okinawa,
Japan.

Innov Clin Neurosci. 2011;8(3):26–33

Funding: There was no funding for this article.

Financial Disclosures: The authors list no conflicts of interest relevant to the content of this article.

Key Words: Alcohol dependence, biological markers, laboratory findings, alcoholism

Abstract: Alcohol dependence is often seen in a variety of clinical settings and requires attention to reduce medical complications, set up appropriate treatments, and minimize utilization of healthcare resources. Patient responses to questionnaires are often used to screen for alcohol problems, but can be misleading in the context of altered mental states or in a patient hesitant to disclose a pattern of alcohol use. Identifying the biochemical consequences of alcohol dependence has led to further study, including correlating laboratory findings to increase accuracy of identifying problem drinkers. Understanding the normal function, mechanism of abnormal findings, sensitivity, and specificity of the current laboratory studies can substantiate clinical suspicion of alcohol use. In this article, we provide results from our literature search regarding laboratory abnormalities in alcohol dependence, review options available to complement a thorough history and physical, and provide a brief overview of future biomarkers for detection of alcohol use.

Introduction

Diagnosing alcohol dependence can be a rewarding opportunity, improving patient’s health and quality of life. Alcohol has detrimental effects, such as contributing to the following: liver disorders, gastrointestinal problems, cardiovascular and diabetic complications, sexual function disorders, birth defects, bone loss, neurological complications, and increased risk of cancer. Alcohol has also been linked to increased motor vehicle accidents and severity of injuries sustained.[38] Despite the significant benefits of addressing alcohol problems with patients, physicians often underdiagnose the severity and prevalence of alcohol dependence.[12] Alcohol biomarkers can either demonstrate the effects of alcohol on the body, classified as indirect markers, or measure alcohol or its metabolites, classified as direct markers. The following are alcohol markers discussed in the literature that continue to be investigated to determine optimal clinical use.

Alanine and Aspartate Aminotransferase

Alanine aminotranferease (ALT) and aspartate aminotransferase (AST), similar to gamma-glutamyl transferase (GGT), indicate generalized hepatic damage.[22] AST is a mitochondrial enzyme found predominantly in the liver, but is also found in the skeletal muscle, heart, pancreas, kidney, brain, lung, and red and white blood cells. Processes affecting muscle, including strenuous exercise, muscle disorders, and many drugs, are the most common nonhepatic cause of an increase in AST.

ALT is cytosolic and mainly present in hepatic tissue, making its elevation slightly more specific to liver injury. Both enzymes have been used to identify liver damage for decades as large elevations are associated with viral hepatitis or toxin-induced injury.[11]

Accuracy of elevated transaminases indicating alcohol-induced injury is low compared to GGT. Most review articles quote a large range for both sensitivity and specificity depending on the study used, ranging from 10 to 90 percent for both.[1,24] Table 1, Part 1 Table 1, Part 2 lists a few of the primary source values encountered. Alcoholic hepatitis has been shown to preferentially elevate AST. An AST/ALT ratio of 2.0 has been accepted to indicate alcohol-induced damage 90 percent of the time (high specificity), but is not necessarily elevated to that proportion in all alcohol-dependent patients (low sensitivity). Transaminases have been shown to be useful in monitoring known alcoholic patients for relapse. A 40-percent increase in AST is 90-percent sensitive, and a 20-percent increase in ALT is 80-percent sensitive for relapse, yet both have low specificity. However, like GGT, since these labs are economical and widely available, they are used often.[11]

Gamma-Glutamyl Transferase

GGT is a membrane-anchored enzyme that functions, as its name suggests, to transfer a glutamyl group onto certain amino acids (or water in glutathione metabolism). Important in liver function, GGT is also found in the spleen, kidneys, pancreas, biliary tree, heart, brain, and seminal vesicles.[15] Most often used along with alkaline phosphatase (ALP) as a marker of biliary stasis, its elevation is often associated with generalized liver damage. It has also recently been shown to be a good marker for potential type 2 diabetes development.[14]

GGT is believed to be elevated in liver damage in response to increased oxidative stress and resultant decrease in glutathione levels. Repeat ethanol exposure also causes cell inflammation and eventual hepatocyte necrosis, resulting in further increases in serum GGT as the enzyme is released from dying cells.[11]

GGT is often touted as an acute marker for alcohol damage, but it suffers from many limitations. Any cause of biliary damage/stasis or direct hepatocyte injury may elevate GGT. Due to its presence in a vast array of tissues, many other forms of tissue damage may cause elevated GGT. It can even be used as a marker of disease severity and prognosis in cardiovascular disease.[35,36] GGT is also limited as a screening tool primarily due to poor sensitivity; many chronic drinkers no longer have an elevated GGT level.[11]

Most resources place sensitivity and specificity of GGT for alcohol abuse at 40 to 80 percent for both.[24] For patients on inpatient service, the test has increased sensitivity, but at the cost of decreased specificity with more comorbid illnesses causing elevations. Still, GGT is overall more accurate than the other traditional markers (e.g., AST, ALT, mean corpuscular volume [MCV]) for alcohol-induced liver damage.[11] Since the test for GGT remains very inexpensive and is conveniently included in a comprehensive metabolic panel (CMP) and liver function test (LFT) panel, it remains the most commonly used marker for indicating acute alcohol-induced liver damage.[4]

Mean Corpuscular Volume

Mean corpuscular volume (MCV) has also long been known to elevate with chronic alcohol consumption. This elevation appears to be secondary to direct morrow-toxic effects of alcohol, not due to vitamin deficiencies as may be expected. Although alcohol does affect folate absorption, a true folate deficiency appears in only 17 percent of chronic alcoholics, while elevated MCV is more common. And with an erythrocyte half life 120 days, MCV is an indicator of more chronic alcohol consumption and is a poor marker of acute ethanol intake or relapse.[11] Interestingly, in combination with AST, a normal lab value virtually rules out delerium tremens in emergency patients.[13]

MCV is not consistently elevated with alcohol consumption, making it a poor screening tool for alcohol abuse. Sensitivity is generally accepted below 50 percent, although different studies have found a wide range of sensitivities. An elevated MCV above the normal range has a very good specificity, often quoted as over 90 percent.[11,20] MCV has also been shown to be more sensitive in women for unknown reasons. Since MCV is another easily obtained study, its use is encouraged when considering chronic alcohol abuse and dependence.[11,24]

Carbohydrate-Deficient Transferrin

Discovered in 1976,[34] carbohydrate-deficient transferrin (CDT) describes specific isoforms of transferrin with fewer sialic acid residues from normal isoforms (primarily tetrasialo-transferrin).[25] CDT evaluation for alcohol abuse measures Tf2 (disialo-Tf) and lesser-sialylated but often undetectable forms.[3] Tf3 (trisialo-Tf) had previously been included, but studies indicated alcohol consumption did not alter the levels leading to false positives. Isoforms are separated using electrophoresis or chromatography.[5] Chronic alcohol consumption, via ethanol or its metabolite acetaldehyde, inhibits glycosylation/sialylation in the golgi apparatus of hepatocytes, although the specific enzyme inhibited remains in debate.[3,9] The result is a decrease in sialylation of transferrin and an increase in relative amounts of CDT isoforms despite normal total transferrin levels.[25] Since total transferrin levels are unchanged, percentage of CDT is often used to describe CDT levels in patients, as it standardizes the value to patients total CDT.[18]

The most important and common factor influencing CDT testing accuracy to detect ethanol abuse/dependence is gender. Healthy women have higher levels of serum CDT compared to men. Some argue that this is due to menstrual-based subclinical iron deficiency causing increased total transferrin, but no correlation has been accepted between CDT and total Tf. In fact, no correlation has been found between CDT and the menstrual cycle, serum estradiol, serum iron, or oral contraceptive pill use.[2,3] This slight baseline elevation decreases the sensitivity of CDT as an ethanol abuse marker in women.

Many other factors may influence CDT testing accuracy. Rare genetic mutations in amino acid sequences affect the ability to separate isoforms, reducing the test accuracy.[3] Age has been shown to reduce accuracy in some studies, and yet have no effect in others. Elevated body mass index (BMI) might blunt ethanol-dose CDT-response curves. Diastolic blood pressure (BP) greater than 90 might amplify the dose-response curve. Smoking may potentiate elevated CDT levels due to increased liver damage. Drinking patterns also affect CDT. For all of these markers listed, more research is necessary to determine significance.[3] However, recent research has not indicated a need to adjust CDT values for any of the listed factors.[6]

CDT levels have also been shown to parallel malnutrition in patients with anorexia nervosa, increasing with disease. Elevated CDT levels are independent of alcohol use, and are not present in patients with bulimia.[30] More research is needed, but this may be a feature that slightly decreases CDT testing specificity.

The half life of elevated CDT as a marker of heavy ethanol use has been estimated at 1.5 to 2 weeks. Functioning similar to hemoglobin A1c (HbA1c) as a marker for blood sugar control, CDT shows changes only with sustained long-term elevations in blood alcohol concentration (BAC). For this reason, single episodes of acute alcohol intoxication do not elevate CDT. Instead, CDT elevation is a measure of the area under the curve for BAC versus time. Eighty grams, or 5 to 6 drinks, of ethanol daily for three weeks showed minimal increases in CDT.[37] If prolonged, however, small amounts of alcohol consumed over months to years will increase CDT. CDT has been shown to be useful in indicating relapse of alcohol consumption in sober alcoholics.[2] Overall, CDT has been shown to be a promising test to identify alcohol dependence.[3,5]

This article presents a collection of tested sensitivities and specificities for CDT from different articles (Table 1, Part 1 Table 1, Part 2). Although early testing showed sensitivities and specificities in the 1980s and 1990s, recent tests tend to show an overall sensitivity and specificity of 60 to 70 percent and 80 to 90 percent, respectively. Most articles show consistently higher accuracy for detecting alcohol dependence via CDT levels compared to the more commonly used GGT or MCV. One test suggests that GGT is a more useful correlate to heavy drinking in women[5] due to the smaller CDT elevation seen in women with alcoholism. Overall, CDT is the most accurate single serum marker for chronic alcohol use readily available.[3]

Ethyl Glucuronide

The majority of ethanol consumed is metabolized via the oxidative pathway in the liver to acetaldehyde then acetate via dehydrogenase enzymes. A small fraction, estimated at approximately 0.05 percent of consumed ethanol, is conjugated in hepatic endoplasmic reticulum via glucuronosyl transferase into ethyl glucuronide (EtG). Serum EtG peaks 2 to 5 hours after ethanol peaks, and is eliminated with a half life of approximately 2 to 3 hours via excretion in the urine. Since it is a direct metabolite of ethanol, its concentration parallels ethanol, just 2 to 5 hours later, so its area under the curve (AUC) is proportional.[29] This delay has caused urinary EtG to be considered for examining recent inpatient alcohol use.[19] EtG may also be measured in infant meconium as a marker of in-utero exposure to ethanol.[26]

Due to its short half life, serum EtG is not usually measured. Urine EtG is more commonly used. EtG can remain in urine for several days after exposure to alcohol. EtG in hair is used as a longer-term marker of ethanol use. Since hair grows at approximately 1cm/month, the amount of time to be analyzed can be adjusted simply by taking more or less hair to analyze, usually up to 6cm.[27] EtG is anionic and hydrophilic, and is transferred into the hair shaft primarily via blood in the hair follicle vasculature. However, EtG excreted in urine can elevate levels in pubic hair due to proximity. Increased amounts of sweat produced in the scalp may also cause increased deposition of EtG in hair. Washing habits may remove EtG from the hair shaft over time, affecting the accuracy of the test.[29]

Sensitivities and specificities have consistently been around 90 percent in both as indicated in Table 1, Part 1 Table 1, Part 2. The payoff is a long, complex, and expensive testing process. This test is still new and concern about nonbeverage sources of alcohol resulting in false positives, specifically ethanol-based hand sanitizing gel that is commonly used in the healthcare setting, has been expressed.[33] More research is indicated to ensure its consistency and to determine other factors that may cause variability in accuracy, such as melanin content or BMI.[27]

Ethyl sulfate (EtS) is another direct biomarker used in detecting alcohol use and can be used in conjunction with EtG to help in detection of recent alcohol use. It is another phase-II reaction during which ethanol is conjugated with sulfate by sulfotransferase.[19]

Combination Tests

As shown in Table 1, Part 1 Table 1, Part 2, many researchers have attempted to improve accuracy of analyzing ethanol dependence via combinations of existing lab markers. Often, combinations include CDT since it has been shown to have the highest accuracy for any single lab test.

CDT + GGT. As mentioned previously, CDT has a poor sensitivity for women, so combining it with GGT helps to alleviate that shortcoming and improve accuracy. This improvement also holds true for men. The result is considered positive if either test is positive alone. This improves the sensitivity with minimal reduction in an already high specificity. Results for some studies using this method are listed in Table 1, Part 1 Table 1, Part 2. Overall return of sensitivity and specificity has been 70 to 90 percent and 60 to 100 percent, respectively.[2,24] This test has been used for multiple purposes, such as assessing risk of reinstating a drivers license in a patient who was previously dependent on alcohol.[7]

The Antilla index (AI) is another method of combining these values using a mathematical formula to weight each test and forming a cutoff for abnormal. This test has proven to be effective at increasing sensitivity without affecting specificity, similar to the more basic combination of CDT and GGT described previously.[2]

CDT + MCV. Using a similar method as described for CDT + GGT, CDT + MCV has also been shown to improve sensitivity without affecting specificity. Calculated values are presented in Table 1, Part 1 Table 1, Part 2.[31]

CDT + early detection of alcohol consumption. Early detection of alcohol consumption (EDAC) is a statistical evaluation of between 10 and 40 common laboratory values, encompassing liver function tests to electrolytes, and comparing the results with profiles from heavy to light drinkers. The results are reported as correlation values. Used alone, the EDAC has demonstrated varying specificities and sensitivities attributed to different research models (use of structured and vastly different alcohol consumption providing optimistic results). However, it may be useful as a screening tool for heavy drinking with CDT used as a confirmatory test.[17]

Other combinations. Many other combinations have been attempted to improve sensitivity and specificity, although none have been accepted as standard of care. A few algorithms are summarized in Table 1, Part 1 Table 1, Part 2.[31] These use a similar mechanism as described for combining earlier tests, where a certain number of abnormal values are required to consider the entire test abnormal. In the larger combinations, CDT and MCV are given twice the weighting of homcysteine, folate, or GGT. The listed combinations do indeed show excellent accuracy, although more research is needed to duplicate results and to examine different possible combinations.

Alcohol Biomarkers for Future Use

Many other tests are being researched as markers of alcohol dependence. Although more research is required to determine accuracy and precision, the following include direct and indirect markers that suggest unique purposes for clinical evaluations.

Serum total sialic acid (TSA) is also elevated in patients with chronic alcoholism for the same reason that CDT is elevated; alcohol causes desialylation of glycoproteins long term. Sensitivity and specificity have been reported at 55 and 93 percent, respectively, for TSA,[9] with no difference for free sialic acid.[10] TSA has the advantage over CDT because it produces quicker results and is more cost effective.

Whole blood associated acetaldehyde (WBAA) is the evaluation of free acetaldehyde and hemoglobin-associated acetaldehyde (HAA) and is useful to measure drinking patterns over time because HAA remains elevated for approximately one month.[28]

Phosphatidylethanol (PEth) is a phospholipid produced only in the presence of ethanol via phospholipase D. It is elevated in as little as three weeks and remains elevated for 14 days after discontinuing alcohol consumption.[16] It provides the advantage of requiring an average daily alcohol consumption of 50g, or 4 to 5 drinks, daily with a total consumption of about 1,000g.

Fatty acid ethyl esters (FAEE), also increase with chronic alcohol consumption, and are deposited in hair with comparable mechanism and sensitivity/specificity to EtG.[29]

Discussion

Biomarkers suggestive of alcohol dependence and drinking patterns remain an important tool for inpatient management and successful outpatient treatment. Traditional laboratory tests, including AST, ALT, GGT, and MCV, suffer from low sensitivities and specificities. However, the tests benefit from standardization and widespread use to assist clinicians in determining treatment course. Continued research to determine optimal use of these tests continues, but alternative biomarkers to chronicle alcohol use patterns are important. Understanding a patient’s alcohol history, such as binge drinking versus chronic alcoholism, could result in differing treatment options, including acute detoxification in individuals presenting with markedly high BAC.

The most popular new test being researched is CDT, with significantly improved sensitivity and specificity over the existing lab tests. This improved accuracy comes with the price of long turnaround time and high cost, similar to hair EtG and FAEE measurements. Current research demonstrates a broad interest in finding new biomarkers in assessing a patient’s alcohol consumption.

Combination testing is also useful to evaluate alcohol use. Using tests in conjunction with one another provides greater sensitivity, but may lower specificity. Obtaining test results also depends on the turnaround time of the longest test. However, using a highly sensitive test with confirmatory testing can provide significant increases in specificity at the cost of decreased sensitivity. CDT has been extensively researched to help correlate effectiveness in conjunction with other commonly used tests.

Table 2 summarizes available specificities and sensitivities of standalone tests and of combination tests. As demonstrated in these values, no consensus has been reached regarding absolute accuracy values of discussed markers. Still, clinical relevance can be taken from these values. CDT use is recommended for accurate determination of alcohol dependence, so long as immediate results are not critical for clinical decision making. For more urgent decisions, traditional laboratory tests supporting clinical judgement remain the recommended course.

Research continues to explore new biomarkers to detect problem-drinking patterns with greater sensitivity and specificity. Current directions include analyzing direct biomarkers for clinical usefulness and determining appropriate indirect biomarkers to assess damage from alcohol use. A particularly useful possible future study would attempt to alleviate some of the discrepancies between accuracy values for the existing tests by using a large population. Interest also lies in using algorithms of common tests to increase accuracy of available tools. A combination of the more traditional tests with fast turnaround would improve accuracy above a single marker for more urgent clinical decisions, and would be another useful future direction. The application of such combination tests should be easier with the advent of electronic medical records (EMR). EMR would allow for an algorithm to automatically calculate the appropriate weighted combination of individual markers for improved accuracy in clinical practice. As research continues to delineate benefits and pitfalls of various tests, clinicians will be able to implement optimized strategies in evaluation of patients with suspicious histories.

References
1. Aertgeerts B, Buntinx F, Ansoms S, Fevery J. Screening properties of questionnaires and laboratory tests for the detection of alcohol abuse or dependence in a general practice population. Br J Gen Pract. 2001;51(464):206–217.
2. Anton RF, Lieber C, Tabakoff B; CDTect Study Group. Carbohydrate-deficient transferrin and g-glutamyltransferase for the detection and monitoring of alcohol use: results from a multisite study. Alcohol Clin Exp Res. 2002;26(8):1215–1222.
3. Arndt T. Carbohydrate-deficient transferrin as a marker of chronic alcohol abuse: a critical review of preanalysis, analysis, and interpretation. Clin Chem. 2001;47(1):13–27.
4. Balldin J, Berggren U, Berglund K, et al. Gamma-glutamyltransferase in alcohol use disorders: modification of decision limits in relation to treatment goals? Scand J Clin Lab Invest. 2010;70(2):71–74.
5. Bergström JP, Helander A. Clinical characteristics of carbohydrate-deficient transferrin (% disialotransferrin) measured by HPLC: sensitivity, specificity, gender effects, and relationship with other alcohol biomarkers. Alcohol Alcohol. 2008;43(4):436–441. Epub 2008 Apr 14.
6. Bergström JP, Helander A. Influence of alcohol use, ethnicity, age, gender, BMI and smoking on the serum transferrin glycoform pattern: implications for use of carbohydrate-deficient transferrin (CDT) as alcohol biomarker. Clin Chim Acta. 2008;388(1-2):59–67. Epub 2007 Oct 13.
7. Bianchi V, Ivaldi A, Raspagni A, et al. Use of carbohydrate-deficient transferrin (CDT) and a combination of GGT and CDT (GGT-CDT) to assess heavy alcohol consumption in traffic medicine. Alcohol Alcohol. 2010;45(3):247–251. Epub 2010 Jan 28.
8. Bortolotti F, De Paoli G, Tagliaro F. Carbohydrate-deficient transferrin (CDT) as a marker of alcohol abuse: a critical review of the literature 2001–2005. J Chromatogr B Analyt Technol Biomed Life Sci. 2006;841(1–2):96–109. Epub 2006 May 24.
9. Chrostek L, Cylwik B, Korcz W, et al. Serum free sialic acid as a marker of alcohol abuse. Alcohol Clin Exp Res. 2007;31(6):996–1001. Epub 2007 Apr 11.
10. Chrostek L, Cylwik B, Krawiec A. Relationship between serum sialic acid and sialylated glycoproteins in alcoholics. Alcohol Alcohol. 2007;42(6):588–592.
11. Conigrave KM, Davies P, Haber P, Whitfield JB. Traditional markers of excessive alcohol use. Addiction. 2003;98 Suppl 2:31–43.
12. Dawes MA, Frank S, Rost K. Clinician assessment of psychiatric comorbidity and alcoholism severity in adult alcoholic inpatients. Am J Drug Alcohol Abuse. 1993;19(3):377–386.
13. Findley JK, Park LT, Siefert CJ, et al. Two routine blood tests-mean corpuscular volume and aspartate aminotransferase-as predictors of delirium tremens in trauma patients. J Trauma. 2010;69(1):199–201.
14. Gautier A, Balkau B, Lange C, et al. Risk factors for incident type 2 diabetes in individuals with a BMI of <27 kg/m2: the role of gamma-glutamyltransferase. Data from an Epidemiological Study on the Insulin Resistance Syndrome (DESIR). Diabetologia. 2010;53(2):247–253. Epub 2009 Nov 20.
15. Goldberg DM. Structural, functional, and clinical aspects of gamma-glutamyltransferase. CCRC Crit Rev Clin Lab Sci. 1980;12(1):1–58.
16. Hartmann S, Aradottir S, Graf M, et al. Phosphatidylethanol as a sensitive and specific biomarker: comparison with gamma-glutamyl transpeptidase, mean corpuscular volume and carbohydrate-deficient transferrin. Addict Biol. 2007;12(1):81–84.
17. Harasymiw J, Bean P. The combined use of the early detection of alcohol consumption (EDAC) test and carbohydrate-deficient transferrin to identify heavy drinking behaviour in males. Alcohol Alcohol. 2001;36(4):349–353.
18. Jeppsson JO, Arndt T, Schellenberg F, et al. Toward standardization of carbohydrate-deficient transferrin (CDT) measurements: I. Analyte definition and proposal of a candidate reference method. Clin Chem Lab Med. 2007;45(4):558–562.
19. Junghanns K, Graf I, Pflüger J, et al. Urinary ethyl glucuronide (EtG) and ethyl sulphate (EtS) assessment: valuable tools to improve verification of abstention in alcohol-dependent patients during in-patient treatment and at follow-ups. Addiction. 2009;104(6):921–926.
20. Koivisto H, Hietala J, Anttila P, et al. Long-term ethanol consumption and macrocytosis: diagnostic and pathogenic implications. J Lab Clin Med. 2006;147(4):191–196.
21. Kravos M, Malesic I. Glutamate dehydrogenase as a marker of alcohol dependence. Alcohol Alcohol. 2010;45(1):39–44. Epub 2009 Oct 7.
22. Loomba R, Bettencourt R, Barrett-Connor E. Synergistic association between alcohol intake and body mass index with serum alanine and aspartate aminotransferase levels in older adults: the rancho bernardo study. Aliment Pharmacol Ther. 2009;30(11-12):1137–1149. Epub 2009 Sep 8.
23. Lovinger DM, Crabbe JC. Laboratory models of alcoholism: treatment target identification and insight into mechanisms. Nat Neurosci. 2005;8(11):1471–1480.
24. Marcos Martín M, Pastor Encinas I, Laso Guzmán F. [Biological markers for alcoholism]. Revista Clínica Española. 2005;205(9):443–445.
25. Mårtensson O, Härlin A, Brandt R, et al. Transferrin isoform distribution: gender and alcohol consumption. Alcohol Clin Exp Res. 1997;21(9):1710–1715.
26. Morini L, Marchei E, Vagnarelli F, et al. Ethyl glucuronide and ethyl sulfate in meconium and hair-potential biomarkers of intrauterine exposure to ethanol. Forensic Sci Int. 2010;196(1–3):74–77. Epub 2010 Jan 8.
27. Morini L, Politi L, Polettini A. Ethyl glucuronide in hair. A sensitive and specific marker of chronic heavy drinking. Addiction. 2009;104(6):915–920. Epub 2009 Apr 9.
28. Peterson, K. (2005). Biomarkers for alcohol use and abuse—a summary. Alcohol Res Health. 2004-2005;28(1):30–37.
29. Pragst F, Rothe M, Moench B, et al. Combined use of fatty acid ethyl esters and ethyl glucuronide in hair for diagnosis of alcohol abuse: interpretation and advantages. Forensic Sci Int. 2010;196(1–3):101–110. Epub 2010 Jan 12.
30. Reif A, Fallgatter AJ, Schmidtke A. Carbohydrate-deficient transferrin parallels disease severity in anorexia nervosa. Psychiatry Res. 2005;137(1-2):143–146. Epub 2005 Oct 20.
31. Rinck D, Frieling H, Freitag A, et al. Combinations of carbohydrate-deficient transferrin, mean corpuscular erythrocyte volume, gamma-glutamyltransferase, homocysteine and folate increase the significance of biological markers in alcohol dependent patients. Drug Alcohol Depend. 2007;89(1):60–65. Epub 2007 Jan 17.
32. Schröder H, de la Torre R, Estruch R, et al. Alcohol consumption is associated with high concentrations of urinary hydroxytyrosol. Am J Clin Nutr. 2009;90(5):1329–1335. Epub 2009 Sep 16.
33. Skipper GE, Wurst F, Weinmann W, Liepman M. Ethanol-based hand sanitizing gel vapor causes positive alcohol marker, ethylglucuronide, and positive breathalyzer. J Addict Med. 2009;3(2):1–5.
34. Stibler H, Kjellin KG. Isoelectric focusing and electrophoresis of the CSF proteins in tremor of different origins. J Neurol Sci. 1976;30(2–3):269–285.
35. Turgut O, Tandogan I, Gurlek A. Association of gamma-glutamyltransferase with cardiovascular risk: a prognostic outlook. Arch Med Res. 2009;40(4):318–320. Epub 2009 Jun 13.
36. Turgut O, Yilmaz A, Yalta K, et al. gamma-Glutamyltransferase is a promising biomarker for cardiovascular risk. Med Hypotheses. 2006;67(5):1060–1064. Epub 2006 Aug 7.
37. Walter H, Hertling I, Benda N, et al. Sensitivity and specificity of carbohydrate-deficient transferrin in drinking experiments and different patients. Alcohol. 2001;25(3):189–194.
38. World Health Organization. Consequences of alcohol use. WHO Global Status Report. 2004:35–88. http://www.who.int/substance_abuse/publications/alcohol/en/index.html Accessed March 23, 2011.