Blood Biomarkers of Alcohol Use: A Scoping Review

Abstract

Purpose of Review

Understanding whether a person has consumed alcohol or not, as well as quantitative assessment of alcohol use, are often based on self-reported measures, which may be subject to recall bias, among other challenges. Although not without limitations, blood biomarkers may complement self-reported assessments to provide a more accurate determination of the presence and quantity of alcohol use. The aim of this review is to provide a critical overview of the current knowledge and research on biomarkers of alcohol use, with a particular focus on blood tests.

Recent Findings

This scoping review summarizes the published work on blood tests currently used in clinical practice, including phosphatidyl ethanol (PEth), fatty acid ethyl ester (FAEE), carbohydrate-deficient transferrin (CDT), total serum sialic acid (TSA), mean corpuscular volume (MCV), alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma glutamyl transpeptidase (GGT), and cholesteryl ester transfer protein (CETP). Emerging blood biomarkers with a potential use to assess alcohol drinking are also briefly reviewed, including N-Acetyl-β-Hexosaminidase (Beta-Hex), macrophage migration inhibitory factor (MIF), and D-dopachrome tautomerase (DDT). We discuss the aforementioned biomarkers in the context of their clinical implications, characteristics, strengths, and limitations.

Summary

The available blood biomarkers considerably vary in the time period in which they detect alcohol use and the amount of alcohol they are sensitive to. While currently available biomarkers provide useful information, especially in combination with self-reported measures, future work is needed to identify more sensitive and specific blood biomarkers for different levels and patterns of alcohol use. Integration of such biomarkers into clinical practice and research will increase the accuracy and richness of the data and may guide more effective and targeted strategies for prevention, diagnosis, and treatment of excessive alcohol use.

Background

Excessive alcohol drinking and alcohol use disorder (AUD) are public health concerns, contributing to a myriad of negative medical, psychosocial, and economic consequences worldwide. Alcohol use, especially excessive drinking, is associated with significant health risks, including mental health problems (e.g., depression and anxiety), medical conditions (e.g., liver and cardiovascular diseases), psychosocial dysfunction, and premature mortality [1–3]. Furthermore, in addition to individual harm, excessive alcohol use is associated with harm to others due to car accidents, domestic violence, and other consequences [4]. Thus, early detection of excessive alcohol use is crucial for preventive, diagnostic, and therapeutic endeavors in this regard.

Clinical assessments, including self-reported measures, are often used to quantify alcohol use. Self-reported measures such as the Alcohol Timeline Followback (TLFB) instruct individuals to recall daily drinking over a specified period of time, using anchors and memory aids to enhance recall [5]. While the TLFB and other self-report measures have shown high validity and reliability, such assessments challenge one’s memory to recall the date, frequency, and amount of alcohol use, with the possibility of introducing recall bias [6]. Comorbid cognitive deficits and/or psychiatric disorders may also limit the accuracy of data collected via retrospective self-reported assessments [7]. Objective and quantitative measures, including blood biomarkers, offer an additional tool to evaluate both acute and long-term alcohol use. In addition to verifying information from self-reported assessments, such biomarkers provide important information about the impact of different amounts, patterns, and timeframes of alcohol use on various organs, tissues, and physiological processes.

The aim of this scoping review is to present a critical overview of the current knowledge and research on blood biomarkers of alcohol use. Although different types of biospecimen, such as the urine, saliva, and hair, have been examined to detect alcohol use [8, 9], the blood is by far the most commonly used sample for laboratory measurements in general practice and primary care settings [10]. In addition, specific to alcohol, biomarkers are generally detectable in the blood earlier and offer a wider range of detection, compared to other biospecimens [11]. While blood biomarkers of alcohol use have been explored in previous research, this scoping review hopes to provide a summary of the most relevant and commonly used markers, with their pros and cons, as well as a roadmap for using blood biomarkers in addiction medicine, particularly as a tool in combination with self-reported assessments. An additional aim is to highlight areas related to each biomarker that warrant further research. Given that the current review is primarily focused on clinical implications of blood biomarkers for alcohol use, for other domains, such as the use of these biomarkers for forensic purposes, technical laboratory details, and their applications in AUD clinical trials, we refer the readers to: [12, 13••, 14••].

Methods

A literature search was performed for the papers published in English on direct and indirect blood biomarkers of alcohol use in the following databases: MEDLINE, Scopus, Embase, and Web of Science. To focus on the most recent literature, the search was limited to the time window from January 2000 to December 2020; however, references of the retrieved records were also screened and papers relevant to the scope of this review were added. The search included four broad terms, adapted to each database: alcoholism, alcohol use disorder/s, alcohol misuse, and biomarker/s. The search strategy was developed, and the literature search was performed with the support of an Informationist at the National Institutes of Health (NIH) Library. EndNote X9 was used to collect, de-duplicate, and organize the records. We included those records that reported development, characteristics, and/or use of blood biomarker for alcohol use. Studies that focused on non-blood (e.g., hair and urine) biomarkers for alcohol use were excluded; for an overview of such non-blood biomarkers, see: [8, 9]. Studies geared toward the use and/or effects of alcohol in specific conditions, such as comorbid health problems (e.g., HIV, heart disease, cancer, obesity), mental health problems (e.g., bipolar disorder, depression, psychosis), or other alcohol-related diseases (e.g., fetal alcohol syndrome, alcohol-associated liver disease) were also excluded. The initial literature search retrieved 2468 records, out of which 866 records were duplicates. Of the unique 1602 records retrieved, 203 records were used to inform the present scoping review; however, not all records are cited in this paper (e.g., review articles, commentaries, or multiple records discussing the same biomarker). The full list of these records can be found in the supplement.

Findings

Direct Blood Biomarkers of Alcohol Use

There are several blood biomarkers that directly examine alcohol use by measuring alcohol levels or byproducts of alcohol metabolism (see Table 1). The use of these biomarkers in clinical settings has been increasing, as they provide direct, valid, and conceivable tools to detect alcohol use and to examine alcohol-related patterns, such as excessive drinking, abstinence, and relapse. The most well-known direct biomarker of alcohol use is Blood Alcohol Concentration (BAC) [11]. In addition to outpatient and ambulatory settings, direct measurement of alcohol concentrations in the blood (and often its proxy, in the breath) is typically used in the emergency room to assess current level of intoxication (e.g., in the context of driving under the influence). Other direct blood biomarkers of alcohol use which are related to alcohol metabolism include phosphatidyl ethanol (PEth) and fatty acid ethyl ester (FAEE). While BAC and FAEEs are biomarkers that have been used in animal research and can be translatable into human research [15, 16], PEth in animals’ blood is not suitable to be compared with clinical work [17]. One advantage of direct biomarkers of alcohol use is that, compared to traditional indirect biomarker, they are less sensitive to comorbid conditions such as liver disease [18]. However, inter-individual variability must be taken into account, as age, sex, and other factors considerably influence alcohol pharmacokinetics [19–23].

Table 1.

Main characteristics of the most common alcohol-related blood biomarkers outlined in this review

Biomarker	Detection window since alcohol exposure	Pattern/amount of alcohol drinking
Phosphatidyl ethanol (PEth)	Up to three weeks after alcohol consumption	50 grams of alcohol per day
Fatty Acid Ethyl Ester (FAEE)	24–99 hours after alcohol consumption	Excessive alcohol use but does not quantitatively correlate with amount of alcohol
Carbohydrate-Deficient Transferrin (CDT)	Up to three weeks after alcohol consumption	5–7 standard drinks per day; recent heavy alcohol use; may be altered due to abstinence
Total Serum Sialic Acid (TSA)	2–5 weeks after alcohol consumption	Excessive alcohol use, or about 60 grams per day of alcohol
Mean Corpuscular Volume (MCV)	Remains high even after several (about 2–4) months of abstinence	Excessive alcohol use, or about 60 grams per day of alcohol
Alanine Aminotransferase (ALT) and Aspartate Aminotransferase (AST)	Elevated after alcohol use for as long as 2–3 weeks	Excessive alcohol intake
Gamma Glutamyl Transpeptidase (GGT)	Between 2–6 weeks after alcohol consumption	Excessive alcohol intake
Cholesteryl Ester Transfer Protein (CETP)	24–48 hours after alcohol consumption	Excessive alcohol intake
N-Acetyl-β-Hexosaminidase (Beta-Hex)	Stays elevated until about 7–10 days after consumption	Can detect acute consumption (120–160 grams of ethanol) and heavy alcohol use
Macrophage Migration Inhibitory Factor (MIF) and D-dopachrome Tautomerase (DDT)	N/A	N/A

Phosphatidyl Ethanol

PEth is a cellular membrane phospholipid that can be measured in the blood as a product of phospholipase D, which catalyzes the reaction between phosphatidylcholine and ethanol [24, 25]. PEth levels can be used to detect heavy alcohol consumption, or about 50 g of alcohol per day, for up to 3 weeks [26–28]. PEth has been shown to have a higher sensitivity for alcohol exposure, compared to other biomarkers, including carbohydrate-deficient transferrin (CDT), gamma glutamyl transpeptidase (GGT), and mean corpuscular volume (MCV), which are discussed below [29]. Studies have examined the validity of PEth to differentiate individuals with current AUD from abstainers and social drinkers and found higher sensitivity and specificity for this marker, compared to other blood biomarkers such as GGT and MCV [30] and fatty acid ethyl ester (FAEEs) in the hair [31]. Most studies to date have tested PEth in whole blood samples as a biomarker of exposure to alcohol [32]. Some research has also tested dried blood spot assays and found similar reliability as PEth in the whole blood [33]. Using dried blood spot assays of PEth may be particularly advantageous due to ease of use, cost-efficiency, and its ability to identify prenatal alcohol exposure in newborns at risk for fetal alcohol-spectrum disorder [34]. Future research, however, is needed to evaluate the application and accuracy of PEth-dried blood spot assays in relation to specific amounts and patterns of alcohol use, for example to distinguish between individuals who are abstinent versus heavy drinkers [35•]. Additionally, future research is necessary to investigate the window of detection and the stability of PEth-dried blood spot assays in comparison with other well-known biomarkers [33, 34].

Fatty Acid Ethyl Ester

FAEE are non-oxidative metabolites of ethanol that are produced from triglycerides or free fatty acids by FAEE synthases and other enzymes [36]. FAEEs are typically measured in plasma and are present in the blood for 24–99 h after alcohol consumption [37], particularly among individuals with chronic excessive alcohol use [38]. It has been suggested that FAEE measurements can detect excessive drinking and exposure to alcohol, but do not quantitatively correlate with the amount of alcohol use [39]. Compared to FAEEs in the blood, measurement of FAEEs via scalp hair is considered a more precise method for detecting exposure to alcohol [13••, 40].

Indirect Blood Biomarkers of Alcohol Use

Indirect biomarkers are indicative of the effects of alcohol on the body chemistry and various organs. A variety of indirect biomarkers have been used to detect and measure exposure to alcohol, including carbohydrate-deficient transferrin (CDT), total serum sialic acid (TSA), mean corpuscular volume (MCV), alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma glutamyl transpeptidase (GGT), and cholesteryl ester transfer protein (CETP) (see Table 1). Moreover, while these biomarkers are typically used in human research as well as in clinical practice, previous evidence has shown that CDT, TSA, MCV, ALT, AST, GGT, and CETP can also be used in animals as indicators of alcohol use [41–44]. While indirect biomarkers of alcohol consumption have been used for decades to estimate the amount, length, and pattern of alcohol intake, it is important to note that factors such as age, sex, time since last alcohol intake, use of other substances, comorbid alcohol-related diseases (e.g., alcohol-associated liver disease), and other medical conditions not related to alcohol may influence the results and must be taken into consideration when interpreting these laboratory tests [31]. For example, serum concentrations of MCV, GGT, AST, and ALT increase with age in individuals 30–70 years old [45]. Thus, these indirect blood biomarkers may be more sensitive in individuals between ages 30 and 70, relative to other age groups, due to normative age-related changes in their blood levels [45, 46].

Carbohydrate-Deficient Transferrin

Carbohydrate-deficient transferrin (CDT) is a type of molecule deficient in carbohydrate sialic acid that carries iron into the bloodstream, known as glycoprotein transferrin. Alcohol drinking disrupts the ability for sialic acid to attach to transferrin [47]. CDT is the most studied and widely used biomarker to detect recent excessive alcohol use. CDT is highly sensitive to heavy alcohol consumption above 40 g per day, or about 5–7 standard drinks per day, and can be used to assess excessive drinking [48, 49]. CDT remains elevated for up to 3 weeks after alcohol consumption and individuals with chronic alcohol use have higher levels of CDT, compared to those who do not drink alcohol [47]. CDT may be particularly valuable in assessing pattern and severity of alcohol use during treatment. For example, when CDT is measured at the start of treatment, a decrease of 30% may be indicative of abstinence, while a 30% increase may indicate relapse [50, 51]. CDT, in combination with GGT, can be even more useful in confirming alcohol use, identifying abstinence and relapse, as well as differentiating individuals with chronic and heavy alcohol use (> 280 g of ethanol per week) from those with lower levels of drinking [52]. Comorbid conditions such hypertension, asthma, depression, and gastrointestinal diseases or concomitant prescription drugs do not influence the specificity of CDT, nor do they elevate CDT levels [53]. More recent assays of CDT have been developed to provide higher sensitivity and specificity, relative to original assays, in the context of conditions that may lead to elevated CDT, such as pregnancy, anemia, and liver disease; however, the interpretation should still be approached with caution in these conditions [54–56]. While CDT values have shown sex-related differences, as CDT levels are higher in females than men [57], using both CDT and GGT leads to increased sensitivity in both males and females, without a loss of specificity [54].

Total Serum Sialic Acid

Total serum sialic acid (TSA) is the sialic acid in the serum attached to carbohydrate chains of glycoproteins, such as transferrin [58]. TSA is elevated in the blood ranging from 2 to 5 weeks after alcohol consumption [59]. TSA levels have been shown to be elevated in patients with AUD, compared to social drinkers [59, 60]. TSA reduces during abstinence and does not appear to be a helpful marker to detect relapse [40].

Mean Corpuscular Volume

Clinicians have used mean corpuscular volume (MCV), the mean volume of red blood cells, as a biomarker for alcohol consumption. Chronic and excessive drinking leads to enlarged red blood cells due to a variety of reasons, including (but not limited to) direct toxic effects of alcohol and its metabolites on both red blood cells and bone marrow, interaction with erythrocyte metabolism, and poor nutrition, including folate and vitamin B12 deficiencies [61]. High MCV can be indicative of excessive alcohol use, or about 60 g per day of alcohol. Research has shown that MCV is slightly elevated in individuals with excessive alcohol use and remains high even after several (about 2–4) months of abstinence [62]. While MCV may indicate excessive drinking, it is neither sensitive nor specific for alcohol use and factors such as age, sex, and pre-existing conditions must be considered in interpretation of the results. Previous research suggests that the sensitivity of MCV for heavy alcohol use is higher among women than men, while its specificity is higher in men than women [22, 63]. It is important to note that some alcohol-related conditions (e.g., liver cirrhosis, malnutrition, and subsequent iron deficiency anemia) are associated with microcytosis [64–66], which along with alcohol-induced macrocytosis, may lead to a condition characterized by two cell populations (dimorphic anemia) with an average calculated MCV within the normal range. On the other hand, non-alcohol-related conditions, such as megaloblastic anemia, may lead to increased MCV independent of alcohol [66, 67]. Collectively, previous work suggests that, as a biomarker of alcohol use, MCV should be complemented with other markers such as GGT or PEth [29].

Alanine Aminotransferase and Aspartate Aminotransferase

Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) are liver-derived enzymes involved in amino acid metabolism and are commonly used as blood biomarkers of heavy alcohol drinking. ALT and AST levels in the blood may be elevated after heavy consumption of alcohol [68] and remain elevated after alcohol use for as long as 2–3 weeks [69, 70]. While ALT and AST are frequently used as indirect biomarkers of alcohol use, these enzymes are direct markers of liver damage and inflammation, and elevated levels indicate subclinical or clinical hepatic dysfunction [71, 72]. Accordingly, research has shown that individuals with heavy alcohol use, but no alcohol-associated liver disease, do not have elevated ALT or AST [73]. Very high levels of ALT and AST (e.g., 500 units per liter) may be a good indicator of alcohol-associated liver disease [73]. Thus, ALT and AST lack diagnostic accuracy for alcohol use and should be used in combination with other blood biomarkers. The ALT/AST ratio, commonly known as De Ritis ratio, may also provide some information for differentiating alcohol-associated liver disease from other causes of hepatic dysfunction and inflammation. For example, in viral hepatitis, ALT is usually higher than AST, while in alcohol-associated liver disease, AST is usually higher than ALT [74]. This ratio, however, has limited sensitivity and specificity, and must be accompanied by other clinical and laboratory assessments.

Gamma Glutamyl Transpeptidase

Gamma glutamyl transpeptidase (GGT) is a glycoprotein enzyme that aids in digestion and is found in the liver, kidney, spleen, and pancreas [75]. GGT is a widely used biomarker for sustained excessive alcohol intake. After heavy alcohol consumption, GGT levels in the blood increase and stay elevated for several weeks (between 2 and 6 weeks) [76, 77]. While measuring GGT is useful for detecting heavy alcohol consumption, as well as relapse, among individuals with AUD [78], specificity is reduced with comorbid medical conditions not related to alcohol (e.g., nonalcoholic liver diseases, nephrotic syndrome, and pancreatitis) [70]. It is also important to note that several medications, such as phenytoin and barbiturates, cause an enzymatic induction of GGT. That said, GGT appears to be more sensitive than ALT and AST in detecting heavy alcohol use [79].

Cholesteryl Ester Transfer Protein

CETP is a plasma glycoprotein that facilitates the transfer of cholesteryl ester from high-density lipoprotein (HDL) [80]. Evidence has shown that CETP concentration is inversely associated with HDL cholesterol; alcohol use reduces plasma concentration of CETP and increases HDL cholesterol levels. Additionally, research has reported lower CETP activity among individuals with heavy alcohol use [81, 82]. Similar to MCV, GGT, AST, and ALT, CETP may be considered a useful clinical tool for assessing alcohol use; however, its utility is limited due to potential confounding factors such as cardiovascular diseases and nutrition deficiencies [83]. Animal models appear to be informative for clinical research on CETP. For example, lipid transfer protein inhibitors, that are induced by alcohol, and reduced CETP activity, have been found in plasma samples from both animals (e.g., rats and pigs) and humans [84].

Emerging Blood Biomarkers of Alcohol Use

N-Acetyl-β-Hexosaminidase

Beta-Hex is a lysosomal enzyme found in most body tissues, and is involved in, for example, metabolizing carbohydrates and gangliosides in liver cells [60]. While research suggests that Beta-Hex may be a biomarker for acute alcohol intoxication in some animal models, its use in preclinical research has been limited to rats with liver cirrhosis and not alcohol use alone [85]. Beta-Hex has been found to be elevated in individuals with heavy alcohol use [60]. Elevated beta-Hex levels return to baseline after 7–10 days of abstinence. Beta-Hex is a highly sensitive and specific biomarker for detecting heavy alcohol use and its blood levels increase after acute ingestion of alcohol (120–160 g of ethanol) or intoxication. However, access to assays for measuring beta-Hex as a blood biomarker for alcohol use is limited, as it is typically used as a diagnostic test for patients with Tay-Sachs disease (a rare neurodegenerative disorder) and laboratory methods including electrophoresis, isoelectric focusing, capillary zone electrophoresis, and ion exchange chromatograph are laborious. Therefore, alcohol researchers and clinicians have little experience using beta-Hex as a diagnostic tool [40, 67]. Future research is needed to investigate various characteristics of beta-Hex as a potential blood biomarker for alcohol use and whether a large-scale implementation would be feasible and cost-effective.

Macrophage Migration Inhibitory Factor and D-dopachrome Tautomerase

Macrophage migration inhibitory factor (MIF) and its homologue, D-dopachrome tautomerase (DDT), are a family of proteins that inhibit the migration of macrophages [86]. In recent studies, MIF and DDT have been shown to be elevated in rodent models of alcohol dependence, as well as among individuals with AUD [87]. Petralia and colleagues demonstrated a trend toward upregulation of transcriptomic expression of the MIF-DDT axis within the blood among individuals who are at risk for developing AUD. Additionally, higher expression of MIF can be detected among individuals with alcohol-associated liver disease. However, future research is needed to investigate the role that MIF and DDT may play in assessing alcohol use and alcohol-related liver disease to parse out their potential utility as a blood biomarker of alcohol use [87].

Conclusions

The goal of this scoping review was to examine current and emerging blood biomarkers that can assist self-reported assessments in identifying and quantifying alcohol use. We summarized the existing literature on several direct and indirect blood biomarkers, each having their own characteristics, strengths, and limitations. Particularly, these blood biomarkers vary in sensitivity and specificity, time intervals, and ability to detect certain dose ranges of alcohol use. In addition to BAC, other direct alcohol biomarkers, such as PEth and FAEEs, which are non-oxidative metabolites of ethanol, are highly sensitive to alcohol exposure [29]. While indirect alcohol biomarkers, such as MCV, AST, ALT, GGT, and CETP, are effective in indicating heavy alcohol use, they are mainly correlated to the impact of chronic alcohol use on the liver and red blood cells, and are greatly influenced by several factors, such as age, sex, and/or organ damage [78]. CDT appears to be the most widely used blood biomarker for alcohol use and has been regarded as one of the most reliable biomarkers for heavy alcohol consumption. It is suggested to use CDT in combination with other biomarkers and interpret the results with caution in the context of accompanying conditions such as pregnancy, anemia, and liver diseases [56].

Greater availability and familiarity with the use of blood biomarkers may contribute to improved accuracy in diagnosis and treatment monitoring for excessive alcohol use, especially when combined with self-reported assessments. Previous research indicates a key role for animal models to identify and validate blood biomarkers for alcohol use, including FAEEs, BAC, CDT, TSA, MCV, ALT, AST, GGT, CETP, MIF, and DDT. However, there remains alternative blood biomarkers that are only measurable and valid in humans, such as PEth and Beta-Hex; therefore, the translatability of findings across preclinical and clinical research should be taken into account. While informative and useful, there are various limitations associated with using blood biomarkers for alcohol use. Firstly, currently used biomarkers are effective in differentiating excessive alcohol use versus abstinence, but do not seem to provide precise quantitative information about the amount of alcohol use. Secondly, these biomarkers are relatively limited in detecting specific patterns of alcohol drinking such as continuous drinking (daily use without binge drinking), frequent drinking (more than three days per week), or episodic drinking (irregular and less frequent drinking with longer periods of no alcohol drinking and occasional binge drinking). Finally, the sensitivity and specificity of blood biomarkers are also influenced by comorbid physical and/or mental health problems, such as metabolic disorders, liver diseases due to other etiologies, cardiovascular diseases, and mood disorders, just to name a few. Thus, laboratory values of these biomarkers must be combined with self-reported measures and interpreted along with history and physical exam to provide a more accurate assessment of alcohol use.

Future Directions

Future research should investigate novel blood biomarkers that may have better specificity and sensitivity in measuring different aspects of alcohol consumption, including excessive alcohol, periods of abstinence, and moderate alcohol use. Other potential biomarkers, such as cytokines, are not included in this review as they are primarily used in research settings, and have inconsistent sensitivity and specificity, while this review focused on the most important blood biomarkers used in the clinic. Additionally, this review does not include redox markers as they are typically used for liver injuries and not alcohol use, per se. Finding new biomarkers that provide quantitative information on patterns of alcohol use may increase clinicians’ ability to detect, treat, and monitor AUD. Currently available blood biomarkers are more specific to alcohol-related organ damage (e.g., alcohol-associated liver disease), rather than risky alcohol use itself—a key aspect that requires more research. On one hand, examination of blood biomarkers in the context of other disorders among patients with AUD may assist in validating the use of these biomarkers across populations with comorbid conditions, such as alcohol-associated liver disease, cardiovascular disease, and/or other alcohol-related health conditions. On the other hand, there is a need to develop and validate more specific biomarkers of alcohol use among mild-to-moderate drinkers who may not have developed other medical comorbidities. While currently available blood biomarkers have the capability of assessing recent alcohol use, there is a need for more sensitive and specific blood biomarkers to measure different levels of alcohol use within different timeframes, as well as biomarkers that can more precisely detect and quantify alcohol use in the presence or absence of comorbid diseases. Thus, it is imperative to investigate combinations of existing biomarkers and to identify more sensitive, specific, easily accessible, and low-cost blood biomarkers that can be used in addition to self-reported assessments to characterize alcohol use.