ORIGINAL ARTICLE
Determination of Ethyl Glucuronide in Hair for Detection of Alcohol Consumption in Patients After Liver Transplantation Hilke Andresen-Streichert, PhD,* Gregor von Rothkirch,† Eik Vettorazzi, MSc,‡ Alexander Mueller, PhD,* Ansgar W. Lohse, MD,§ Dorothea Frederking, MD,† Barbara Seegers, MD,† Bjoern Nashan, MD,† and Martina Sterneck, MD†§
Background: Early detection of alcohol misuse in orthotopic liver transplantation recipients is essential to offer patients support and prevent organ damage. Here, ethyl glucuronide, a metabolite of ethanol found in hair (hEtG), was evaluated for detection of alcohol consumption.
Methods: In 104 transplant recipients, 31 with underlying alcoholic liver disease (ALD) and 73 with non-ALD, hEtG was determined in addition to the alcohol markers urine EtG, blood ethanol, methanol, and carbohydrate-deficient transferrin. Results were compared with patients’ self-reports in a questionnaire and with physicians’ assessments.
Results: By physicians’ assessments, 22% of the patients were suspected of consuming alcohol regularly, although only 6% of the patients acknowledged consumption of a moderate or high amount of alcohol. By testing all markers except for hEtG, alcohol consumption was detected in 7% of the patients. When hEtG testing was added to the assessment, consumption was detected in 17% of the patients. Hair-EtG determination alone revealed chronic alcohol consumption of .10 g/d in 15% of the patients. ALD patients had a positive hEtG result significantly more often than non-ALD patients did (32% versus 8%; P = 0.003). Also, the concentration of hEtG was higher in ALD patients (P = 0.049) and revealed alcohol abuse with consumption of .60 g ethanol per day in 23% of ALD and 3% of non-ALD patients. Patients’ self-reports and physicians’ assessments had a low sensitivity of 27% and 67%, respectively, for detecting regular alcohol intake as indicated by hEtG.
Conclusions: Hair-EtG determination improved the detection of liver transplant patients who used alcohol, and revealed regular alcohol consumption in 32% of ALD and 8% of non-ALD patients.
Received for publication August 5, 2014; accepted October 28, 2014. From the *Department of Legal Medicine; †Department of Hepatobiliary and Transplant Surgery; ‡Department of Medical Biometry and Epidemiology; and §Department of Gastroenterology, University Medical Center Hamburg—Eppendorf, Hamburg, Germany. H. Andresen-Streichert and G. von Rothkirch have contributed equally to this work. The authors declare no conflict of interest. Correspondence: Martina Sterneck, MD, Department of Hepatobiliary and Transplant Surgery, University Medical Center Hamburg—Eppendorf, Martinistraße 52, 20246 Hamburg, Germany (e-mail: [email protected]). Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
Ther Drug Monit Volume 37, Number 4, August 2015
Key Words: alcoholic liver disease, CDT, hair ethyl glucuronide, methanol, urine ethyl glucuronide (Ther Drug Monit 2015;37:539–545)
INTRODUCTION Alcoholic recidivism after orthotopic liver transplantation (OLT) is a concern because it occurs in approximately 10%–50% of liver transplant patients.1,2 Although mild-tomoderate alcohol drinking may have no direct toxic effects on liver function, it has been found to result in reduced patient compliance, is associated with an increased graft rejection rate,3 and significant reduction of the patient survival rate.1 It is well known that patients’ self-reports of their alcohol consumption are unreliable; therefore, objective and accurate indicators to monitor abstinence are needed.4 Direct detection of ethanol (EtOH) in blood or breath assesses only the alcohol intake during the preceding 10–12 hours. Carbohydrate-deficient transferrin (CDT) is a reliable indirect marker that reflects harmful alcohol intake during the previous 1–4 weeks.5 However, daily consumption of 60–80 g of ethanol for at least 7–10 days is required for obtaining a positive test result. As a congener alcohol, methanol (MeOH) may accumulate in body fluids in cases of continuous heavy drinking, and reflect long-term alcohol intake. After cessation of drinking, the MeOH concentration returns to endogenous levels within 1–2 days.6 The determination of ethyl glucuronide (EtG), a metabolic product of ethanol that is detectable in the urine or hair of patients, offers a new, reliable way to test for alcohol intake. Urinary EtG (uEtG) can be found for up to 80 hours after alcohol consumption and allows the detection of even small amounts of ethanol ingested (,5 g).7,8 A high sensitivity and specificity (89% and 99%, respectively) of uEtG tests in a cohort of OLT candidates and recipients were demonstrated by our group.9 Piano et al10 recently recommended using this marker routinely in liver transplant candidates and recipients to improve the detection of alcohol consumption. Detection of EtG in the scalp hair (hEtG) of patients is a powerful tool for monitoring abstinence over a retrospective time period of up to 6 months. Each hair segment of a 1-cm length reflects alcohol consumption over approximately 1 month. However, if
539
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Andresen-Streichert et al
samples ,3 cm are used, the results should be interpreted with caution,11 because an incorporation of EtG from sweat into hair after recent alcohol consumption is a serious concern.12 Several studies showed a high correlation between daily alcohol intake and hEtG concentrations,13,14 and internationally accepted cut-off values have been defined.11 Due to the high specificity and sensitivity of the detection method, hEtG testing has gained growing interest over the past few years, especially for evaluating alcohol abuse in forensic settings.15 In this prospective study, hEtG testing was assessed for the first time in OLT recipients. The results of the hairEtG test were evaluated along with the results of the analysis of other markers of alcohol consumption, that is, uEtG, CDT, ethanol, and methanol, with patients’ self-reports in a written questionnaire, and with the assessments of the physicians.
MATERIALS AND METHODS Patients The patient population included all OLT recipients who underwent liver transplantation $6 months previously and who presented to the outpatient unit of the University Medical Center Hamburg Eppendorf between January 2012 and March 2012. Patients with severe renal dysfunction [glomerular filtration rate ,30 mL/min] were excluded. The study was approved by the local ethics committee (no. PV4015), and all patients gave written informed consent for their participation in the study.
Alcohol Consumption Assessments Patients were asked to complete a standardized, written questionnaire to disclose their alcohol consumption within the preceding 6 months. The quantity of alcoholic drinks consumed was classified into 5 categories: none, occasional, low (;1 drink per week), moderate (2–6 drinks per week), and high consumption ($7 drinks per week). All patients were evaluated by their transplant physician, and a nonstructured interview was conducted. Subsequently, the physicians were asked to give an assessment of the patient’s alcohol drinking behavior without having access to the results of the patient’s questionnaire or laboratory tests. There were 2 categories: the patient is suspected or is not suspected to consume a moderate to high amount of alcohol.
Ther Drug Monit Volume 37, Number 4, August 2015
repeated alcohol consumption (.10 g ethanol per day); (3) .30 pg/mg: high positive result indicating excessive chronic consumption (consumption of an average of $60 g ethanol per day). EtG in urine, and EtOH, MeOH, and CDT in serum were determined as described previously,9 using cut-off values of $0.5 mg/L, $0.1 g/kg, $5 mg/L, and .2.6%, respectively. Hair-ETG was determined by a validated procedure using liquid chromatography/tandem mass spectrometry after solvent extraction with ultrasonic treatment, according to Suesse et al.15 Briefly, approximately 50 mg of the sample was exactly weighed and successively washed with 3 mL acetone and 3 mL water. The washings were removed, and the samples were cut into small pieces (1–2 mm) after adding 1 mL distilled water, 1 mL acetonitrile, and 25 mL methanol. The internal standard solution (EtG-d5 in water) was added, and the extraction was performed by 4 hours of ultrasonication at 508C. After centrifugation (5600g for 5 minutes), the extract was separated and evaporated to dryness at 808C under nitrogen. The residue was reconstituted in 100 mL mobile phase, and 10 mL was injected for measurement by using liquid chromatography/tandem mass spectrometry. EtG was identified by monitoring 2 precursor-to-product transitions (m/z 221 / 57 and m/z 221 / 55), the transitions m/z 221 / 75 (EtG) and m/z 226 / 75 (EtG-d5) were used for quantification. Limit of detection and limit of quantification were determined according to current guidelines16 with 1 and 2.6 pg/mg, respectively.
Determination of Biochemical Markers Bilirubin, international normalized ratio, mean corpuscular volume (MCV), gamma glutamyltransferase (GGT), alanine aminotransferase (ALT), and aspartate aminotransferase (AST) were analyzed at the time of hair sampling. All retrospective serum creatinine test results that were available within 3 months before hair sampling were pooled, and the mean values were used to calculate the glomerular filtration rate according to Cockcroft-Gault.
Statistical Analysis Fisher exact test and Student t-test were used to compare categorical and continuous variables, respectively, between hEtG-positive and hEtG-negative patients and between results of patients with underlying alcoholic liver disease (ALD) and non-ALD, respectively (IBM SPSS Statistics 19; IBM, Ehningen, Germany).
Determination of Alcohol Consumption Markers For hEtG determination, scalp hair samples were collected by cutting a 0.5-cm thick hair strand close to the skin. The proximal 3-cm hair segment and—if available—the distal 3- to 6-cm hair segments were analyzed for EtG reflecting alcohol consumption 0–3 and 3–6 months before sampling, respectively. In accordance with international standards, the following cut-off values for hEtG were applied11: (1) ,7 pg/ mg: negative result indicating abstainers or rare drinking; (2) 7–30 pg/mg: positive result, which strongly suggests
540
RESULTS Patient Population and Characteristics Of the 130 OLT recipients presenting within the study period, 26 were excluded because of an incomplete patient questionnaire (n = 5), a missing urine sample (n = 7), a missing blood sample (n = 1), or an unusable short hair sample (n = 13). Of the 104 patients included in the study, 31 (30%) patients were diagnosed with underlying ALD and 73 (70%) patients were diagnosed with non-ALD (Table 1). Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Ther Drug Monit Volume 37, Number 4, August 2015
Hair EtG in Patients After Liver Transplantation
TABLE 1. Characteristics of Patients With ALD Versus Non-ALD Total, n = 104
Non-ALD, n = 73
ALD, n = 31
P
64 (62) 56.5 6 12.2 83 6 67.7 0.8 6 0.5 87.3 6 6.3 36.7 6 43.1 (17.0, 25.0, 36.3) 38.1 6 41.9 (15.8, 26.0, 45.3) 130.4 6 175.5 (29.8, 61.5, 164.3) 1.4 6 0.4 66.9 6 25.6 63 (61)
41 (56) 54.5 6 13 89 6 73.9 0.7 6 0.6 86.8 6 6.1 37.7 6 42.5 (19.0, 26.0, 37.0) 39.4 6 41.1 (17.0, 27.0, 47.0) 113.6 6 129 (28.0, 58.0, 161.0) 1.3 6 0.4 67.3 6 27.5 39 (53)
23 (74) 61.2 6 8.5 69 6 47 0.8 6 0.4 88.3 6 6.6 34.4 6 44.4 (14.5, 21.0, 27.5) 34.8 6 43.3 (14.0, 18.0, 37.0) 169.8 6 248.9 (30.0, 85.0, 167.0) 1.5 6 0.5 65.9 6 20.2 24 (77)
0.12 0.003 0.11 0.77 0.31 0.73 0.62 0.25 0.11 0.77 0.03
Patients Male sex, n (%) Age, yrs Time post OLT, mos. Bilirubin, mg/dL MCV, fL AST, U/L ALT, U/L GGT, U/L Creatinine, mg/dL GFR, mL/min Therapy with cyclosporine, n (%)
Data are presented as mean 6 SD. For AST, ALT, and GGT the first quartile (Q1), the median (Q2), and the third quartile (Q3) are given in brackets. P values refer to ALD patients versus non-ALD patients. All laboratory parameters were determined on the day of presentation except for the creatinine and GFR value: Here, the mean of all values available within the preceding 3 months is given. GFR was calculated according to Cockcroft-Gault. GFR, glomerular filtration rate.
Patients’ Questionnaires In the patient questionnaire, 28% (29/104) of the patients reported alcohol consumption within the preceding 6 months, but only 6% (6/104) of the patients acknowledged consumption of a moderate or high amount of alcohol (Table 2). ALD patients reported consumption of a larger amount of alcohol than non-ALD patients did, although the proportion of patients
who acknowledged any alcohol consumption was not different between the patient groups (Table 2).
Physicians’ Assessments Physicians suspected that 22% (23/104) of the patients drank a moderate to high amount of alcohol. Patients with ALD were significantly more often suspected to consume
TABLE 2. Detection of Alcohol Consumption in ALD and Non-ALD Patients Patients Alcohol marker, n (%) Any alcohol marker positive Proximal or distal hEtG . 7 pg/mg* Proximal hEtG . 7 pg/mg† Proximal hEtG $ 30 pg/mg‡ uEtG $ 0.5 mg/L§ CDT $ 2.6%¶ EtOH . 0.1 g/kg MeOH $ 5 mg/Lk Physician’s assessment, n (%) Moderate to high consumption suspected Self-report, n (%) No alcohol Any alcohol Occasional consumption Low consumption Moderate consumption High consumption
Total, n = 104 18 16 15 9 7 2 1 0
(17) (15) (14) (9) (7) (2) (1) (0)
Non-ALD, n = 73 7 6 5 2 1 0 0 0
(10) (8) (7) (3) (1) (0) (0) (0)
ALD, n = 31 11 10 10 7 6 2 1 0
P
(35) (32) (32) (23) (19) (6) (3) (0)
0.002 0.003 0.002 0.003 0.003 0.087 0.298 —
23 (22)
12 (16)
11 (35)
0.041
75 29 21 2 3 3
54 19 15 2 2 0
21 10 6 0 1 3
0.633 0.633 1 0.577 1 0.025
(72) (28) (20) (2) (3) (3)
(74) (26) (20) (3) (3) (0)
(68) (32) (19) (0) (3) (10)
Alcohol consumption was evaluated by the determination of the alcohol markers, physicians’ assessments, and patients’ self-reports in questionnaires. P values refer to the comparison of patients with non-ALD and ALD. *Indicates alcohol consumption of .10 g/d in the preceding 3 and/or 3–6 months. †Indicates alcohol consumption of .10 g/d in the preceding 3 months. ‡Indicates alcohol abuse (.60 g/d) during the preceding 3 months. §Indicates recent alcohol uptake up to 3 days before sampling (no conclusion can be drawn to the amount of alcohol that has been consumed). ¶Reflects harmful alcohol intake (.60 g/d) in the previous 1–4 weeks. kAccumulates during continuous heavy drinking, returns to endogenous levels within 1–2 days.
Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
541
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Andresen-Streichert et al
alcohol regularly as compared with patients with non-ALD (ALD, 35% versus non-ALD, 16%; P = 0.04, Table 2).
Alcohol Consumption Markers Seventeen percent (18/104) of the patients tested positive for at least 1 alcohol marker. ALD patients had a positive marker significantly more often than did non-ALD patients (35% versus 10%, P = 0.002; Table 2). Hair-EtG was positive in all but 2 of the 18 (89%) patients with positive alcohol markers. It was either found as the only positive marker (n = 11) or in combination with other positive alcohol markers (n = 5). In 2 cases, uEtG was the only positive marker that indicated recent but not necessarily chronic alcohol intake (no. 14 and 101; Table 3).
EtG in Hair
In the proximal hair segments, hEtG .7 pg/mg was detected in 15/104 (14%) patients. This indicated regular alcohol consumption of on average .10 g per day within the preceding 3 months. In 60% (9/15) of these patients, the hEtG concentration was .30 pg/mg, which indicates excessive alcohol intake of .60 g ethanol per day during that time (Tables 2 and 3). Distal hair segments were available from 50 patients. The hEtG results from the distal hair segments were similar to those from the corresponding proximal hair segments. A similar proportion of patients (8/50; 16%) was found to have a positive distal hEtG result, and in half of these cases (4/8), the hEtG concentration was very high (.30 pg/mg). Patients with underlying ALD had a positive result for hEtG significantly more often than did non-ALD patients
542
Ther Drug Monit Volume 37, Number 4, August 2015
(Table 2). According to the EtG result from the proximal hair segments, 32% of ALD patients, but only 7% of non-ALD patients, had consumed alcohol regularly within the preceding 3 months (P = 0.002). In addition, the mean EtG concentration in the proximal hair segments was significantly higher in ALD as compared with that in non-ALD patients (mean: 248.6 versus 25.1 pg/ mg; P = 0.049). In 23% (7/31) of ALD patients, but only 3% (2/73) of non-ALD patients, a high concentration of hEtG (.30 pg/mg) revealed excessive chronic consumption and predicted a daily consumption of .60 g ethanol (P = 0.003) (Table 2). Comparison of the characteristics of hEtG-positive and hEtG-negative patients did not reveal any statistically significant differences (Table 4). Eighty-seven percent (13/15) of hEtGpositive patients had an elevated GGT level (.38 U/L), whereas only 2 of 38 patients with a normal GGT were hEtG positive (P = 0.084). The diagnostic sensitivity of an elevated GGT value for a positive hEtG result was 87%, but the diagnostic specificity was only 40% (Table 5). Notably, treatment with Cyclosporine A had no impact on the hEtG result (Table 4).
Common Alcohol Consumption Markers A positive EtOH result was found in only 1 patient (no. 26), who acknowledged regular excessive alcohol consumption and who also had a high concentration of hEtG (Table 3). CDT testing revealed heavy alcohol consumption (.60 g per day) during the preceding 4 weeks in 2 patients (no. 97 and 99), both with underlying ALD. Alcohol abuse was also confirmed in these 2 patients by a positive hEtG and uEtG result. One of these patients reported a high amount of alcohol consumption
Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Ther Drug Monit Volume 37, Number 4, August 2015
Hair EtG in Patients After Liver Transplantation
TABLE 4. Comparison of Patients With a Positive and a Negative hEtG Result in the Proximal Hair Strain Patients Male sex, n (%) Age, yrs Time post OLT, mo Bilirubin, mg/dL MCV, fL AST, U/L ALT, U/L GGT, U/L Creatinine, mg/dL GFR, mL/min Therapy with cyclosporine, n (%)
Total, n = 104
hEtG neg, n = 89
hEtG Pos, n = 15
P
64 (62) 56.5 6 12.2 83 6 67.7 0.8 6 0.5 87.3 6 6.3 36.7 6 43.1 (17.0, 25.0, 36.3) 38.1 6 41.9 (15.8, 26.0, 45.3) 130.4 6 175.5 (29.8, 61.5, 164.3) 1.4 6 0.4 66.9 6 25.6 63 (61)
52 (58) 56.4 6 12.1 89 6 70.7 0.7 6 0.5 86.9 6 6.1 36.5 6 44.9 (17.0, 24.0, 35.0) 38.8 6 44.4 (15.0, 26.0, 46.0) 109.8 6 133.8 (28.0, 54.0, 138.0) 1.3 6 0.4 66.5 6 25.6 52 (58)
12 (80) 56.9 6 13.1 68 6 41.9 0.9 6 0.5 89.3 6 6.9 38 6 30.6 (18.5, 25.0, 46.5) 33.7 6 21.4 (18.5, 26.0, 42.0) 252.4 6 299.9 (89.5, 113.0, 300.5) 1.6 6 0.5 69 6 25.2 11 (73)
0.15 0.89 0.18 0.18 0.24 0.88 0.5 0.1 0.1 0.74 0.39
Data are presented as mean 6 SD. For AST, ALT, and GGT, the first quartile (Q1), the median (Q2), and the third quartile (Q3) are given in brackets. P values refer to hEtGpositive patients versus hEtG-negative patients. All laboratory parameters were determined on the day of presentation except for the creatinine and GFR value: Here, the mean of all the values available within the preceding 3 months is given. GFR was calculated according to Cockcroft-Gault. GFR, glomerular filtration rate.
in the questionnaire; the other patient acknowledged heavy consumption only after confrontation with the test results. None of the patients tested positive for MeOH, using 5 mg/L as the threshold for a positive test result9,17; therefore, this marker was not considered to be a useful indicator of alcohol consumption in our study. In the case of lowering the cut-off value for a positive MeOH test result to 3 mg/L—as used in some other institutions18—MeOH results would become positive in 6 cases. However, no other alcohol marker, self-report, or physician’s assessment pointed to alcohol intake in 4 of these 6 patients. This indicated potentially false positive results. Seven patients (7%) tested positive for uEtG (Table 3). This indicated recent, but possibly low, alcohol intake. Three patients acknowledged alcohol intake in the preceding days, compatible with the uEtG results, whereas 2 patients only acknowledged alcohol intake (no. 97 and 101) after viewing their test results. The other 2 uEtG-positive patients, who reported no alcohol consumption (no. 5 and 6), had a high concentration of EtG in the proximal hair segment, which indicates high alcohol intake.
Diagnostic Value of Alcohol Consumption Markers Using the common alcohol consumption markers EtOH, MeOH, and CDT, 3% of patients were found to consume TABLE 5. The Diagnostic Sensitivity and Diagnostic Specificity of the Patient’s Self-Report, the Physician’s Assessment, and an Elevated GGT for Detecting a Positive Proximal hEtG Result Are Given Patient’s self-report High consumption Moderate or high consumption Physician suspected moderate to high consumption Elevated GGT (.38 U/L)
Sensitivity, %
Specificity, %
20 27 67
100 98 85
87
40
Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
alcohol. When urine EtG concentrations were also considered, recent alcohol intake was detected in an additional 4% of patients. Implementation of the hEtG analysis further increased the detection of alcohol intake, and an additional 10% of the patients tested positive for alcohol consumption.
Correlation Between Patients’ Self-Reports and Alcohol Consumption Markers The patient’s statement of consuming a moderate or high amount of alcohol had a high diagnostic specificity of 98%, but only a low diagnostic sensitivity of 27% for predicting a positive hEtG result (Table 5). All 3 patients who acknowledged consumption of a high amount of alcohol had a positive hEtG result (Table 3). In addition, 43% (10/23) of patients who acknowledged only occasional or a low amount of alcohol consumption had a positive hEtG result, which reveals that their actual alcohol intake was much higher than reported. Five of these patients (no. 6, 10, 51, 75, 97) had hEtG concentrations .30 pg/mg, which indicated chronic alcohol consumption of .60 g/d. The other 5 patients (no. 20, 29, 67, 69, 86) had hEtG concentrations between 7 and 30 pg/mg, which aligns with daily consumption of 10 g–60 g of alcohol. Further, 2 of 75 (3%) patients (no. 5, 89), who did not acknowlede any alcohol consumption, had a positive hEtG result. In summary, selfreports underestimate alcohol consumption among liver transplant patients.
Correlation Between Physicians’ Assessments and Alcohol Consumption Markers The physicians’ assessments had a diagnostic specificity of 85%, but a diagnostic sensitivity and positive predictive value (PPV) of only 67% and 44%, respectively, for predicting regular alcohol intake as reflected by a positive hEtG result (Table 5). Forty-three percent (n = 10/23) of the patients suspected by their physicians to consume a moderate or high amount of alcohol had a positive hEtG result; however, 7% (6/81) of patients not suspected by their physicians of
543
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Andresen-Streichert et al
consuming alcohol regularly also had a positive hEtG result [4 of these patients had large concentrations of EtG in their hair (.30 pg/mg)].
DISCUSSION Despite the requirement that liver transplant recipients with underlying ALD should abstain from drinking alcohol, 32% of the ALD patients were found to have consumed regularly .10 g ethanol daily. Even if the quantity of alcohol consumed does not immediately impair graft function, the negative impact of alcohol consumption on long-term outcome has been reported.1 Methods for the objective detection of alcohol intake in transplant recipients have been limited so far because patient visits to the outpatient clinic are scheduled several weeks in advance, and random testing of alcohol markers is not usually possible. In this study, in addition to the traditional alcohol markers, the concentration of EtG in the hair was determined. Hair-EtG is a particularly useful marker in the transplant setting, because a positive result detects regular alcohol consumption of a clinically relevant quantity in the preceding 3 (proximal hair segment) and 3–6 (distal hair segment) months. Previous studies confirm the high reliability of hEtG testing with a specificity between 94% and 100%.15,19–21 A meta-analysis of data from 8 studies found an overall hEtG sensitivity of 96% and specificity of 99% to detect an alcohol consumption of .60 g/d.22 In the pretransplant setting, our group found that specificity, sensitivity, PPV, and negative predictive value for hEtG testing were 98%, 86%, 92%, and 86%, respectively.23 Stewart et al24 confirmed the utility of hEtG testing for detecting moderate-to-heavy drinking in patients with liver disease. EtG analysis in hair is not influenced by age, sex, body mass index, tobacco smoke, hair color, or the treatment of hair with alcohol-containing cosmetic products.11,15,25,26 Bleaching, perming, and hair dyeing may decrease but not increase the amount of hEtG.15,27,28 With regard to the analytical method, sample preparation seems to influence the sensitivity of EtG detection. In general, but especially when using smaller hair samples, the extraction recovery can be significantly improved by grinding of the hair instead of cutting it.29 Most importantly, only very few patients with analytical false positive hEtG results have been reported in the literature.30–32 The positive results are suggested to be caused by treatment of the hair with EtG-containing hair lotions based on ethanolic plant extracts. Suesse et al15 found 3% “interpretive false positive hEtG results” using a cut-off of 30 pg/mL in an investigation of parents involved in child custody cases. Also, a small pilot study with 12 patients has recently suggested that the concentration of hEtG may be higher than expected among patients with severe renal dysfunction.33 However, in a previous study, we found no correlation between serum creatinine and hEtG,23 but we nevertheless excluded patients with renal insufficiency stages 4 and 5 from this posttransplantation study population. Here, the detection of alcohol consumption was markedly improved among transplant recipients by implementation
544
Ther Drug Monit Volume 37, Number 4, August 2015
of the hEtG analysis. In comparison with determination of all serum and urinary alcohol markers together, including uEtG, CDT, MeOH, and EtOH, determination of hEtG more than doubled the detection rate of alcohol consumption from 7% to 17%. In 2 patients, hEtG was negative, whereas uEtG was positive. This revealed recent alcohol consumption, but only of a small amount. Determination of the EtG concentration in the distal hair segments did not add much relevant information to the result of hEtG in the proximal segment. In a single patient, only the distal, but not the proximal hair segment, was positive for hEtG. This revealed that the patient had reduced regular alcohol consumption at least 3 months before testing. The physicians’ assessments of their patients and the patients’ self-reports were not reliable methods to predict alcohol consumption among OLT patients. The physicians’ assessments had a PPV of only 44% for detecting a clinically meaningful amount of alcohol consumption as indicated by a positive hEtG result. In the patients’ self-reports, only 6% of all study participants acknowledged regular alcohol consumption of a moderate or high amount within the preceding 3 months, but as many as 14% of all transplant recipients had a positive hEtG result obtained from proximal hair segments. Also, 43% and 3% of the patients who acknowledged only occasional or no alcohol consumption, respectively, had positive hEtG results. Notably, based on a positive proximal hEtG result, ALD patients were significantly more often found to consume alcohol regularly as compared with non-ALD patients (P = 0.002). Although the number of ALD patients (n = 31) included in the study was small, the positive hEtG results found among 32% of the ALD patients stand in opposition to their self-reports and provide a good rationale for screening transplant recipients with underlying ALD for alcohol consumption. A high concentration of hEtG .30 pg/mg revealed harmful drinking habits in 23% of all the ALD patients. Further, 7% of the 73 non-ALD patients had a positive hEtG result (Table 2). GGT was more often elevated among hEtG-positive patients than among hEtG-negative patients, indicating the harmful effect of alcohol on graft function. But the PPV of an elevated GGT for detecting patients with chronic alcohol consumption was only 20%.
CONCLUSIONS In conclusion, hEtG determination is a valuable tool for revealing regular alcohol consumption among liver transplant recipients. Patients’ self-reports underestimate the amount of alcohol consumed and physicians’ assessments were also not very reliable. To identify patients early who may benefit from professional alcohol counseling, we now have implemented hEtG testing in the annual routine check-up of all liver transplant recipients with ALD and of all non-ALD patients who are suspected of alcohol consumption by their physician. REFERENCES 1. Pfitzmann R, Schwenzer J, Rayes N, et al. Long-term survival and predictors of relapse after orthotopic liver transplantation for alcoholic liver disease. Liver Transpl. 2007;13:197–205.
Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Ther Drug Monit Volume 37, Number 4, August 2015
2. DiMartini A, Day N, Dew M, et al. Alcohol consumption patterns and predictors of use following liver transplantation for alcoholic liver disease. Liver Transpl. 2006;12:813–820. 3. Cuadrado A, Fábrega E, Casafont F, et al. Alcohol recidivism impairs long-term patient survival after orthotopic liver transplantation for alcoholic liver disease. Liver Transpl. 2005;11:420–426. 4. DiMartini A, Dew M. Monitoring alcohol use on the liver transplant wait list: therapeutic and practical issues. Liver Transpl. 2012;18:1267–1269. 5. Stibler H. Carbohydrate-deficient transferrin in serum: a new marker of potentially harmful alcohol consumption reviewed. Clin Chem. 1991;37: 2029–2037. 6. Helander A, Eriksson C. Laboratory tests for acute alcohol consumption: results of the WHO/ISBRA Study on State and Trait Markers of Alcohol Use and Dependence. Alcohol Clin Exp Res. 2002;26:1070–1077. 7. Thierauf A, Halter CC, Rana S, et al. Urine tested positive for ethyl glucuronide after trace amounts of ethanol. Addiction. 2009;104:2007– 2012. 8. Wurst FM, Kempter C, Seidl S, et al. Ethyl glucuronide—a marker of alcohol consumption and a relapse marker with clinical and forensic implications. Alcohol Alcohol. 1999;34:71–77. 9. Staufer K, Andresen H, Vettorazzi E, et al. Urinary ethyl glucuronide as a novel screening tool in patients pre- and post-liver transplantation improves detection of alcohol consumption. Hepatology. 2011;54: 1640–1649. 10. Piano S, Marchioro L, Gola E, et al. Assessment of alcohol consumption in liver transplant candidates and recipients: the best combination of the tools available. Liver Transpl. 2014;20:815–822. 11. SOHT. 2014 Consensus for the use of alcohol markers in hair for assessment of both abstinence and chronic excessive alcohol consumption. Available at: http://www.soht.org/images/pdf/2014%20Alcohol%20markers %20revision%2013JUN14%20FINAL.pdf. Accessed September 10, 2014. 12. Agius R, Ferreira L, Yegles M. Can ethyl glucuronide in hair be determined only in 3 cm hair strands? Forensic Sci Int. 2012;218:3–9. 13. Appenzeller BM, Agirman R, Neuberg P, et al. Segmental determination of ethyl glucuronide in hair: a pilot study. Forensic Sci Int. 2007; 173:87–92. 14. Politi L, Morini L, Leone F, et al. Ethyl glucuronide in hair: is it a reliable marker of chronic high levels of alcohol consumption? Addiction. 2006; 101:1408–1412. 15. Suesse S, Pragst F, Mieczkowski T, et al. Practical experiences in application of hair fatty acid ethyl esters and ethyl glucuronide for detection of chronic alcohol abuse in forensic cases. Forensic Sci Int. 2012;218:82–91. 16. Paul L, Mußhoff F, Aebi B, et al. Richtlinie der GTFCh zur Qualitätssicherung bei forensisch-toxikologischen Untersuchungen (guidelines for quality assurance in forensic-toxicological analyses) [in German]. Available at: http://wwwgtfchorg/cms/images/stories/files/GTFCh_Richtlinie_ For-Tox_Version%201pdf. Accessed September 6, 2009. 17. Huckenbeck W, Bonte W. Alkohologie. In: Madea B, Brinkmann B, eds. Handbuch Gerichtliche Medizin. Heidelberg, Germany: Springer; 2003. 18. Hempel J, Greif-Higer G, Kaufmann T, et al. Detection of alcohol consumption in patients with alcoholic liver cirrhosis during the
Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
Hair EtG in Patients After Liver Transplantation
19.
20. 21.
22.
23.
24. 25. 26. 27. 28. 29. 30. 31. 32. 33.
evaluation process for liver transplantation. Liver Transpl. 2012;18: 1310–1315. Kharbouche H, Sporkert F, Troxler S, et al. Development and validation of a gas chromatography-negative chemical ionization tandem mass spectrometry method for the determination of ethyl glucuronide in hair and its application to forensic toxicology. J Chromatogr B Analyt Technol Biomed Life Sci. 2009;877:2337–2343. Lees R, Kingston R, Williams T, et al. Comparison of ethyl glucuronide in hair with self-reported alcohol consumption. Alcohol Alcohol. 2012; 47:267–272. Hastedt M, Büchner M, Rothe M, et al. Detecting alcohol abuse: traditional blood alcohol markers compared to ethyl glucuronide (EtG) and fatty acid ethyl esters (FAEEs) measurement in hair. Forensic Sci Med Pathol. 2013;9:184–193. Boscolo-Berto R, Favretto D, Cecchetto G, et al. Sensitivity and specificity of EtG in hair as a marker of chronic excessive drinking: pooled analysis of raw data and meta-analysis of diagnostic accuracy studies. Ther Drug Monit 2014;36:560–567. Sterneck M, Yegles M, von Rothkirch G, et al. Determination of ethyl glucuronide in hair improves evaluation of long-term alcohol abstention in liver transplant candidates. Liver Int. 2013. doi: 10.1111/liv.12243. Epub ahead of print. Stewart S, Koch D, Willner I, et al. Hair ethyl glucuronide is highly sensitive and specific for detecting moderate-to-heavy drinking in patients with liver disease. Alcohol Alcohol. 2013;48:83–87. Gareri J, Appenzeller B, Walasek P, et al. Impact of hair-care products on FAEE hair concentrations in substance abuse monitoring. Anal Bioanal Chem. 2011;400:183–188. Martins Ferreira L, Binz T, Yegles M. The influence of ethanol containing cosmetics on ethyl glucuronide concentration in hair. Forensic Sci Int. 2012;218:123–125. Yegles M, Labarthe A, Auwarter V, et al. Comparison of ethyl glucuronide and fatty acid ethyl ester concentrations in hair of alcoholics, social drinkers and teetotallers. Forensic Sci Int. 2004;145:167–173. Morini L, Politi L, Polettini A. Ethyl glucuronide in hair. A sensitive and specific marker of chronic heavy drinking. Addiction. 2009;104:915–920. Crunelle C, Cappelle D, Covaci A, et al. Hair ethyl glucuronide as a biomarker of alcohol consumption in alcohol-dependent patients: role of gender differences. Drug Alcohol Depend. 2014;141:163–166. Sporkert F, Kharbouche H, Augsburger MP, et al. Positive EtG findings in hair as a result of a cosmetic treatment. Forensic Sci Int. 2012;218: 97–100. Kintz P, Nicholson D. Testing for ethanol markers in hair: discrepancies after simultaneous quantification of ethyl glucuronide and fatty acid ethyl esters. Forensic Sci Int. 2014;13:44–46. Kintz P, Nicholson D. Interpretation of a highly positive ethyl glucuronide together with negative fatty acid ethyl esters in hair and negative blood results. Forensic Toxicol. 2014;32:176–179. Høiseth G, Morini L, Ganss R, et al. Higher levels of hair ethyl glucuronide in patients with decreased kidney function. Alcohol Clin Exp Res. 2013;37(suppl 1):E14–E16.
545
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Determination of Ethyl Glucuronide in Hair for Detection of Alcohol Consumption in Patients After Liver Transplantation Hilke Andresen-Streichert, PhD,* Gregor von Rothkirch,† Eik Vettorazzi, MSc,‡ Alexander Mueller, PhD,* Ansgar W. Lohse, MD,§ Dorothea Frederking, MD,† Barbara Seegers, MD,† Bjoern Nashan, MD,† and Martina Sterneck, MD†§
Background: Early detection of alcohol misuse in orthotopic liver transplantation recipients is essential to offer patients support and prevent organ damage. Here, ethyl glucuronide, a metabolite of ethanol found in hair (hEtG), was evaluated for detection of alcohol consumption.
Methods: In 104 transplant recipients, 31 with underlying alcoholic liver disease (ALD) and 73 with non-ALD, hEtG was determined in addition to the alcohol markers urine EtG, blood ethanol, methanol, and carbohydrate-deficient transferrin. Results were compared with patients’ self-reports in a questionnaire and with physicians’ assessments.
Results: By physicians’ assessments, 22% of the patients were suspected of consuming alcohol regularly, although only 6% of the patients acknowledged consumption of a moderate or high amount of alcohol. By testing all markers except for hEtG, alcohol consumption was detected in 7% of the patients. When hEtG testing was added to the assessment, consumption was detected in 17% of the patients. Hair-EtG determination alone revealed chronic alcohol consumption of .10 g/d in 15% of the patients. ALD patients had a positive hEtG result significantly more often than non-ALD patients did (32% versus 8%; P = 0.003). Also, the concentration of hEtG was higher in ALD patients (P = 0.049) and revealed alcohol abuse with consumption of .60 g ethanol per day in 23% of ALD and 3% of non-ALD patients. Patients’ self-reports and physicians’ assessments had a low sensitivity of 27% and 67%, respectively, for detecting regular alcohol intake as indicated by hEtG.
Conclusions: Hair-EtG determination improved the detection of liver transplant patients who used alcohol, and revealed regular alcohol consumption in 32% of ALD and 8% of non-ALD patients.
Received for publication August 5, 2014; accepted October 28, 2014. From the *Department of Legal Medicine; †Department of Hepatobiliary and Transplant Surgery; ‡Department of Medical Biometry and Epidemiology; and §Department of Gastroenterology, University Medical Center Hamburg—Eppendorf, Hamburg, Germany. H. Andresen-Streichert and G. von Rothkirch have contributed equally to this work. The authors declare no conflict of interest. Correspondence: Martina Sterneck, MD, Department of Hepatobiliary and Transplant Surgery, University Medical Center Hamburg—Eppendorf, Martinistraße 52, 20246 Hamburg, Germany (e-mail: [email protected]). Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
Ther Drug Monit Volume 37, Number 4, August 2015
Key Words: alcoholic liver disease, CDT, hair ethyl glucuronide, methanol, urine ethyl glucuronide (Ther Drug Monit 2015;37:539–545)
INTRODUCTION Alcoholic recidivism after orthotopic liver transplantation (OLT) is a concern because it occurs in approximately 10%–50% of liver transplant patients.1,2 Although mild-tomoderate alcohol drinking may have no direct toxic effects on liver function, it has been found to result in reduced patient compliance, is associated with an increased graft rejection rate,3 and significant reduction of the patient survival rate.1 It is well known that patients’ self-reports of their alcohol consumption are unreliable; therefore, objective and accurate indicators to monitor abstinence are needed.4 Direct detection of ethanol (EtOH) in blood or breath assesses only the alcohol intake during the preceding 10–12 hours. Carbohydrate-deficient transferrin (CDT) is a reliable indirect marker that reflects harmful alcohol intake during the previous 1–4 weeks.5 However, daily consumption of 60–80 g of ethanol for at least 7–10 days is required for obtaining a positive test result. As a congener alcohol, methanol (MeOH) may accumulate in body fluids in cases of continuous heavy drinking, and reflect long-term alcohol intake. After cessation of drinking, the MeOH concentration returns to endogenous levels within 1–2 days.6 The determination of ethyl glucuronide (EtG), a metabolic product of ethanol that is detectable in the urine or hair of patients, offers a new, reliable way to test for alcohol intake. Urinary EtG (uEtG) can be found for up to 80 hours after alcohol consumption and allows the detection of even small amounts of ethanol ingested (,5 g).7,8 A high sensitivity and specificity (89% and 99%, respectively) of uEtG tests in a cohort of OLT candidates and recipients were demonstrated by our group.9 Piano et al10 recently recommended using this marker routinely in liver transplant candidates and recipients to improve the detection of alcohol consumption. Detection of EtG in the scalp hair (hEtG) of patients is a powerful tool for monitoring abstinence over a retrospective time period of up to 6 months. Each hair segment of a 1-cm length reflects alcohol consumption over approximately 1 month. However, if
539
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Andresen-Streichert et al
samples ,3 cm are used, the results should be interpreted with caution,11 because an incorporation of EtG from sweat into hair after recent alcohol consumption is a serious concern.12 Several studies showed a high correlation between daily alcohol intake and hEtG concentrations,13,14 and internationally accepted cut-off values have been defined.11 Due to the high specificity and sensitivity of the detection method, hEtG testing has gained growing interest over the past few years, especially for evaluating alcohol abuse in forensic settings.15 In this prospective study, hEtG testing was assessed for the first time in OLT recipients. The results of the hairEtG test were evaluated along with the results of the analysis of other markers of alcohol consumption, that is, uEtG, CDT, ethanol, and methanol, with patients’ self-reports in a written questionnaire, and with the assessments of the physicians.
MATERIALS AND METHODS Patients The patient population included all OLT recipients who underwent liver transplantation $6 months previously and who presented to the outpatient unit of the University Medical Center Hamburg Eppendorf between January 2012 and March 2012. Patients with severe renal dysfunction [glomerular filtration rate ,30 mL/min] were excluded. The study was approved by the local ethics committee (no. PV4015), and all patients gave written informed consent for their participation in the study.
Alcohol Consumption Assessments Patients were asked to complete a standardized, written questionnaire to disclose their alcohol consumption within the preceding 6 months. The quantity of alcoholic drinks consumed was classified into 5 categories: none, occasional, low (;1 drink per week), moderate (2–6 drinks per week), and high consumption ($7 drinks per week). All patients were evaluated by their transplant physician, and a nonstructured interview was conducted. Subsequently, the physicians were asked to give an assessment of the patient’s alcohol drinking behavior without having access to the results of the patient’s questionnaire or laboratory tests. There were 2 categories: the patient is suspected or is not suspected to consume a moderate to high amount of alcohol.
Ther Drug Monit Volume 37, Number 4, August 2015
repeated alcohol consumption (.10 g ethanol per day); (3) .30 pg/mg: high positive result indicating excessive chronic consumption (consumption of an average of $60 g ethanol per day). EtG in urine, and EtOH, MeOH, and CDT in serum were determined as described previously,9 using cut-off values of $0.5 mg/L, $0.1 g/kg, $5 mg/L, and .2.6%, respectively. Hair-ETG was determined by a validated procedure using liquid chromatography/tandem mass spectrometry after solvent extraction with ultrasonic treatment, according to Suesse et al.15 Briefly, approximately 50 mg of the sample was exactly weighed and successively washed with 3 mL acetone and 3 mL water. The washings were removed, and the samples were cut into small pieces (1–2 mm) after adding 1 mL distilled water, 1 mL acetonitrile, and 25 mL methanol. The internal standard solution (EtG-d5 in water) was added, and the extraction was performed by 4 hours of ultrasonication at 508C. After centrifugation (5600g for 5 minutes), the extract was separated and evaporated to dryness at 808C under nitrogen. The residue was reconstituted in 100 mL mobile phase, and 10 mL was injected for measurement by using liquid chromatography/tandem mass spectrometry. EtG was identified by monitoring 2 precursor-to-product transitions (m/z 221 / 57 and m/z 221 / 55), the transitions m/z 221 / 75 (EtG) and m/z 226 / 75 (EtG-d5) were used for quantification. Limit of detection and limit of quantification were determined according to current guidelines16 with 1 and 2.6 pg/mg, respectively.
Determination of Biochemical Markers Bilirubin, international normalized ratio, mean corpuscular volume (MCV), gamma glutamyltransferase (GGT), alanine aminotransferase (ALT), and aspartate aminotransferase (AST) were analyzed at the time of hair sampling. All retrospective serum creatinine test results that were available within 3 months before hair sampling were pooled, and the mean values were used to calculate the glomerular filtration rate according to Cockcroft-Gault.
Statistical Analysis Fisher exact test and Student t-test were used to compare categorical and continuous variables, respectively, between hEtG-positive and hEtG-negative patients and between results of patients with underlying alcoholic liver disease (ALD) and non-ALD, respectively (IBM SPSS Statistics 19; IBM, Ehningen, Germany).
Determination of Alcohol Consumption Markers For hEtG determination, scalp hair samples were collected by cutting a 0.5-cm thick hair strand close to the skin. The proximal 3-cm hair segment and—if available—the distal 3- to 6-cm hair segments were analyzed for EtG reflecting alcohol consumption 0–3 and 3–6 months before sampling, respectively. In accordance with international standards, the following cut-off values for hEtG were applied11: (1) ,7 pg/ mg: negative result indicating abstainers or rare drinking; (2) 7–30 pg/mg: positive result, which strongly suggests
540
RESULTS Patient Population and Characteristics Of the 130 OLT recipients presenting within the study period, 26 were excluded because of an incomplete patient questionnaire (n = 5), a missing urine sample (n = 7), a missing blood sample (n = 1), or an unusable short hair sample (n = 13). Of the 104 patients included in the study, 31 (30%) patients were diagnosed with underlying ALD and 73 (70%) patients were diagnosed with non-ALD (Table 1). Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Ther Drug Monit Volume 37, Number 4, August 2015
Hair EtG in Patients After Liver Transplantation
TABLE 1. Characteristics of Patients With ALD Versus Non-ALD Total, n = 104
Non-ALD, n = 73
ALD, n = 31
P
64 (62) 56.5 6 12.2 83 6 67.7 0.8 6 0.5 87.3 6 6.3 36.7 6 43.1 (17.0, 25.0, 36.3) 38.1 6 41.9 (15.8, 26.0, 45.3) 130.4 6 175.5 (29.8, 61.5, 164.3) 1.4 6 0.4 66.9 6 25.6 63 (61)
41 (56) 54.5 6 13 89 6 73.9 0.7 6 0.6 86.8 6 6.1 37.7 6 42.5 (19.0, 26.0, 37.0) 39.4 6 41.1 (17.0, 27.0, 47.0) 113.6 6 129 (28.0, 58.0, 161.0) 1.3 6 0.4 67.3 6 27.5 39 (53)
23 (74) 61.2 6 8.5 69 6 47 0.8 6 0.4 88.3 6 6.6 34.4 6 44.4 (14.5, 21.0, 27.5) 34.8 6 43.3 (14.0, 18.0, 37.0) 169.8 6 248.9 (30.0, 85.0, 167.0) 1.5 6 0.5 65.9 6 20.2 24 (77)
0.12 0.003 0.11 0.77 0.31 0.73 0.62 0.25 0.11 0.77 0.03
Patients Male sex, n (%) Age, yrs Time post OLT, mos. Bilirubin, mg/dL MCV, fL AST, U/L ALT, U/L GGT, U/L Creatinine, mg/dL GFR, mL/min Therapy with cyclosporine, n (%)
Data are presented as mean 6 SD. For AST, ALT, and GGT the first quartile (Q1), the median (Q2), and the third quartile (Q3) are given in brackets. P values refer to ALD patients versus non-ALD patients. All laboratory parameters were determined on the day of presentation except for the creatinine and GFR value: Here, the mean of all values available within the preceding 3 months is given. GFR was calculated according to Cockcroft-Gault. GFR, glomerular filtration rate.
Patients’ Questionnaires In the patient questionnaire, 28% (29/104) of the patients reported alcohol consumption within the preceding 6 months, but only 6% (6/104) of the patients acknowledged consumption of a moderate or high amount of alcohol (Table 2). ALD patients reported consumption of a larger amount of alcohol than non-ALD patients did, although the proportion of patients
who acknowledged any alcohol consumption was not different between the patient groups (Table 2).
Physicians’ Assessments Physicians suspected that 22% (23/104) of the patients drank a moderate to high amount of alcohol. Patients with ALD were significantly more often suspected to consume
TABLE 2. Detection of Alcohol Consumption in ALD and Non-ALD Patients Patients Alcohol marker, n (%) Any alcohol marker positive Proximal or distal hEtG . 7 pg/mg* Proximal hEtG . 7 pg/mg† Proximal hEtG $ 30 pg/mg‡ uEtG $ 0.5 mg/L§ CDT $ 2.6%¶ EtOH . 0.1 g/kg MeOH $ 5 mg/Lk Physician’s assessment, n (%) Moderate to high consumption suspected Self-report, n (%) No alcohol Any alcohol Occasional consumption Low consumption Moderate consumption High consumption
Total, n = 104 18 16 15 9 7 2 1 0
(17) (15) (14) (9) (7) (2) (1) (0)
Non-ALD, n = 73 7 6 5 2 1 0 0 0
(10) (8) (7) (3) (1) (0) (0) (0)
ALD, n = 31 11 10 10 7 6 2 1 0
P
(35) (32) (32) (23) (19) (6) (3) (0)
0.002 0.003 0.002 0.003 0.003 0.087 0.298 —
23 (22)
12 (16)
11 (35)
0.041
75 29 21 2 3 3
54 19 15 2 2 0
21 10 6 0 1 3
0.633 0.633 1 0.577 1 0.025
(72) (28) (20) (2) (3) (3)
(74) (26) (20) (3) (3) (0)
(68) (32) (19) (0) (3) (10)
Alcohol consumption was evaluated by the determination of the alcohol markers, physicians’ assessments, and patients’ self-reports in questionnaires. P values refer to the comparison of patients with non-ALD and ALD. *Indicates alcohol consumption of .10 g/d in the preceding 3 and/or 3–6 months. †Indicates alcohol consumption of .10 g/d in the preceding 3 months. ‡Indicates alcohol abuse (.60 g/d) during the preceding 3 months. §Indicates recent alcohol uptake up to 3 days before sampling (no conclusion can be drawn to the amount of alcohol that has been consumed). ¶Reflects harmful alcohol intake (.60 g/d) in the previous 1–4 weeks. kAccumulates during continuous heavy drinking, returns to endogenous levels within 1–2 days.
Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
541
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Andresen-Streichert et al
alcohol regularly as compared with patients with non-ALD (ALD, 35% versus non-ALD, 16%; P = 0.04, Table 2).
Alcohol Consumption Markers Seventeen percent (18/104) of the patients tested positive for at least 1 alcohol marker. ALD patients had a positive marker significantly more often than did non-ALD patients (35% versus 10%, P = 0.002; Table 2). Hair-EtG was positive in all but 2 of the 18 (89%) patients with positive alcohol markers. It was either found as the only positive marker (n = 11) or in combination with other positive alcohol markers (n = 5). In 2 cases, uEtG was the only positive marker that indicated recent but not necessarily chronic alcohol intake (no. 14 and 101; Table 3).
EtG in Hair
In the proximal hair segments, hEtG .7 pg/mg was detected in 15/104 (14%) patients. This indicated regular alcohol consumption of on average .10 g per day within the preceding 3 months. In 60% (9/15) of these patients, the hEtG concentration was .30 pg/mg, which indicates excessive alcohol intake of .60 g ethanol per day during that time (Tables 2 and 3). Distal hair segments were available from 50 patients. The hEtG results from the distal hair segments were similar to those from the corresponding proximal hair segments. A similar proportion of patients (8/50; 16%) was found to have a positive distal hEtG result, and in half of these cases (4/8), the hEtG concentration was very high (.30 pg/mg). Patients with underlying ALD had a positive result for hEtG significantly more often than did non-ALD patients
542
Ther Drug Monit Volume 37, Number 4, August 2015
(Table 2). According to the EtG result from the proximal hair segments, 32% of ALD patients, but only 7% of non-ALD patients, had consumed alcohol regularly within the preceding 3 months (P = 0.002). In addition, the mean EtG concentration in the proximal hair segments was significantly higher in ALD as compared with that in non-ALD patients (mean: 248.6 versus 25.1 pg/ mg; P = 0.049). In 23% (7/31) of ALD patients, but only 3% (2/73) of non-ALD patients, a high concentration of hEtG (.30 pg/mg) revealed excessive chronic consumption and predicted a daily consumption of .60 g ethanol (P = 0.003) (Table 2). Comparison of the characteristics of hEtG-positive and hEtG-negative patients did not reveal any statistically significant differences (Table 4). Eighty-seven percent (13/15) of hEtGpositive patients had an elevated GGT level (.38 U/L), whereas only 2 of 38 patients with a normal GGT were hEtG positive (P = 0.084). The diagnostic sensitivity of an elevated GGT value for a positive hEtG result was 87%, but the diagnostic specificity was only 40% (Table 5). Notably, treatment with Cyclosporine A had no impact on the hEtG result (Table 4).
Common Alcohol Consumption Markers A positive EtOH result was found in only 1 patient (no. 26), who acknowledged regular excessive alcohol consumption and who also had a high concentration of hEtG (Table 3). CDT testing revealed heavy alcohol consumption (.60 g per day) during the preceding 4 weeks in 2 patients (no. 97 and 99), both with underlying ALD. Alcohol abuse was also confirmed in these 2 patients by a positive hEtG and uEtG result. One of these patients reported a high amount of alcohol consumption
Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Ther Drug Monit Volume 37, Number 4, August 2015
Hair EtG in Patients After Liver Transplantation
TABLE 4. Comparison of Patients With a Positive and a Negative hEtG Result in the Proximal Hair Strain Patients Male sex, n (%) Age, yrs Time post OLT, mo Bilirubin, mg/dL MCV, fL AST, U/L ALT, U/L GGT, U/L Creatinine, mg/dL GFR, mL/min Therapy with cyclosporine, n (%)
Total, n = 104
hEtG neg, n = 89
hEtG Pos, n = 15
P
64 (62) 56.5 6 12.2 83 6 67.7 0.8 6 0.5 87.3 6 6.3 36.7 6 43.1 (17.0, 25.0, 36.3) 38.1 6 41.9 (15.8, 26.0, 45.3) 130.4 6 175.5 (29.8, 61.5, 164.3) 1.4 6 0.4 66.9 6 25.6 63 (61)
52 (58) 56.4 6 12.1 89 6 70.7 0.7 6 0.5 86.9 6 6.1 36.5 6 44.9 (17.0, 24.0, 35.0) 38.8 6 44.4 (15.0, 26.0, 46.0) 109.8 6 133.8 (28.0, 54.0, 138.0) 1.3 6 0.4 66.5 6 25.6 52 (58)
12 (80) 56.9 6 13.1 68 6 41.9 0.9 6 0.5 89.3 6 6.9 38 6 30.6 (18.5, 25.0, 46.5) 33.7 6 21.4 (18.5, 26.0, 42.0) 252.4 6 299.9 (89.5, 113.0, 300.5) 1.6 6 0.5 69 6 25.2 11 (73)
0.15 0.89 0.18 0.18 0.24 0.88 0.5 0.1 0.1 0.74 0.39
Data are presented as mean 6 SD. For AST, ALT, and GGT, the first quartile (Q1), the median (Q2), and the third quartile (Q3) are given in brackets. P values refer to hEtGpositive patients versus hEtG-negative patients. All laboratory parameters were determined on the day of presentation except for the creatinine and GFR value: Here, the mean of all the values available within the preceding 3 months is given. GFR was calculated according to Cockcroft-Gault. GFR, glomerular filtration rate.
in the questionnaire; the other patient acknowledged heavy consumption only after confrontation with the test results. None of the patients tested positive for MeOH, using 5 mg/L as the threshold for a positive test result9,17; therefore, this marker was not considered to be a useful indicator of alcohol consumption in our study. In the case of lowering the cut-off value for a positive MeOH test result to 3 mg/L—as used in some other institutions18—MeOH results would become positive in 6 cases. However, no other alcohol marker, self-report, or physician’s assessment pointed to alcohol intake in 4 of these 6 patients. This indicated potentially false positive results. Seven patients (7%) tested positive for uEtG (Table 3). This indicated recent, but possibly low, alcohol intake. Three patients acknowledged alcohol intake in the preceding days, compatible with the uEtG results, whereas 2 patients only acknowledged alcohol intake (no. 97 and 101) after viewing their test results. The other 2 uEtG-positive patients, who reported no alcohol consumption (no. 5 and 6), had a high concentration of EtG in the proximal hair segment, which indicates high alcohol intake.
Diagnostic Value of Alcohol Consumption Markers Using the common alcohol consumption markers EtOH, MeOH, and CDT, 3% of patients were found to consume TABLE 5. The Diagnostic Sensitivity and Diagnostic Specificity of the Patient’s Self-Report, the Physician’s Assessment, and an Elevated GGT for Detecting a Positive Proximal hEtG Result Are Given Patient’s self-report High consumption Moderate or high consumption Physician suspected moderate to high consumption Elevated GGT (.38 U/L)
Sensitivity, %
Specificity, %
20 27 67
100 98 85
87
40
Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
alcohol. When urine EtG concentrations were also considered, recent alcohol intake was detected in an additional 4% of patients. Implementation of the hEtG analysis further increased the detection of alcohol intake, and an additional 10% of the patients tested positive for alcohol consumption.
Correlation Between Patients’ Self-Reports and Alcohol Consumption Markers The patient’s statement of consuming a moderate or high amount of alcohol had a high diagnostic specificity of 98%, but only a low diagnostic sensitivity of 27% for predicting a positive hEtG result (Table 5). All 3 patients who acknowledged consumption of a high amount of alcohol had a positive hEtG result (Table 3). In addition, 43% (10/23) of patients who acknowledged only occasional or a low amount of alcohol consumption had a positive hEtG result, which reveals that their actual alcohol intake was much higher than reported. Five of these patients (no. 6, 10, 51, 75, 97) had hEtG concentrations .30 pg/mg, which indicated chronic alcohol consumption of .60 g/d. The other 5 patients (no. 20, 29, 67, 69, 86) had hEtG concentrations between 7 and 30 pg/mg, which aligns with daily consumption of 10 g–60 g of alcohol. Further, 2 of 75 (3%) patients (no. 5, 89), who did not acknowlede any alcohol consumption, had a positive hEtG result. In summary, selfreports underestimate alcohol consumption among liver transplant patients.
Correlation Between Physicians’ Assessments and Alcohol Consumption Markers The physicians’ assessments had a diagnostic specificity of 85%, but a diagnostic sensitivity and positive predictive value (PPV) of only 67% and 44%, respectively, for predicting regular alcohol intake as reflected by a positive hEtG result (Table 5). Forty-three percent (n = 10/23) of the patients suspected by their physicians to consume a moderate or high amount of alcohol had a positive hEtG result; however, 7% (6/81) of patients not suspected by their physicians of
543
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Andresen-Streichert et al
consuming alcohol regularly also had a positive hEtG result [4 of these patients had large concentrations of EtG in their hair (.30 pg/mg)].
DISCUSSION Despite the requirement that liver transplant recipients with underlying ALD should abstain from drinking alcohol, 32% of the ALD patients were found to have consumed regularly .10 g ethanol daily. Even if the quantity of alcohol consumed does not immediately impair graft function, the negative impact of alcohol consumption on long-term outcome has been reported.1 Methods for the objective detection of alcohol intake in transplant recipients have been limited so far because patient visits to the outpatient clinic are scheduled several weeks in advance, and random testing of alcohol markers is not usually possible. In this study, in addition to the traditional alcohol markers, the concentration of EtG in the hair was determined. Hair-EtG is a particularly useful marker in the transplant setting, because a positive result detects regular alcohol consumption of a clinically relevant quantity in the preceding 3 (proximal hair segment) and 3–6 (distal hair segment) months. Previous studies confirm the high reliability of hEtG testing with a specificity between 94% and 100%.15,19–21 A meta-analysis of data from 8 studies found an overall hEtG sensitivity of 96% and specificity of 99% to detect an alcohol consumption of .60 g/d.22 In the pretransplant setting, our group found that specificity, sensitivity, PPV, and negative predictive value for hEtG testing were 98%, 86%, 92%, and 86%, respectively.23 Stewart et al24 confirmed the utility of hEtG testing for detecting moderate-to-heavy drinking in patients with liver disease. EtG analysis in hair is not influenced by age, sex, body mass index, tobacco smoke, hair color, or the treatment of hair with alcohol-containing cosmetic products.11,15,25,26 Bleaching, perming, and hair dyeing may decrease but not increase the amount of hEtG.15,27,28 With regard to the analytical method, sample preparation seems to influence the sensitivity of EtG detection. In general, but especially when using smaller hair samples, the extraction recovery can be significantly improved by grinding of the hair instead of cutting it.29 Most importantly, only very few patients with analytical false positive hEtG results have been reported in the literature.30–32 The positive results are suggested to be caused by treatment of the hair with EtG-containing hair lotions based on ethanolic plant extracts. Suesse et al15 found 3% “interpretive false positive hEtG results” using a cut-off of 30 pg/mL in an investigation of parents involved in child custody cases. Also, a small pilot study with 12 patients has recently suggested that the concentration of hEtG may be higher than expected among patients with severe renal dysfunction.33 However, in a previous study, we found no correlation between serum creatinine and hEtG,23 but we nevertheless excluded patients with renal insufficiency stages 4 and 5 from this posttransplantation study population. Here, the detection of alcohol consumption was markedly improved among transplant recipients by implementation
544
Ther Drug Monit Volume 37, Number 4, August 2015
of the hEtG analysis. In comparison with determination of all serum and urinary alcohol markers together, including uEtG, CDT, MeOH, and EtOH, determination of hEtG more than doubled the detection rate of alcohol consumption from 7% to 17%. In 2 patients, hEtG was negative, whereas uEtG was positive. This revealed recent alcohol consumption, but only of a small amount. Determination of the EtG concentration in the distal hair segments did not add much relevant information to the result of hEtG in the proximal segment. In a single patient, only the distal, but not the proximal hair segment, was positive for hEtG. This revealed that the patient had reduced regular alcohol consumption at least 3 months before testing. The physicians’ assessments of their patients and the patients’ self-reports were not reliable methods to predict alcohol consumption among OLT patients. The physicians’ assessments had a PPV of only 44% for detecting a clinically meaningful amount of alcohol consumption as indicated by a positive hEtG result. In the patients’ self-reports, only 6% of all study participants acknowledged regular alcohol consumption of a moderate or high amount within the preceding 3 months, but as many as 14% of all transplant recipients had a positive hEtG result obtained from proximal hair segments. Also, 43% and 3% of the patients who acknowledged only occasional or no alcohol consumption, respectively, had positive hEtG results. Notably, based on a positive proximal hEtG result, ALD patients were significantly more often found to consume alcohol regularly as compared with non-ALD patients (P = 0.002). Although the number of ALD patients (n = 31) included in the study was small, the positive hEtG results found among 32% of the ALD patients stand in opposition to their self-reports and provide a good rationale for screening transplant recipients with underlying ALD for alcohol consumption. A high concentration of hEtG .30 pg/mg revealed harmful drinking habits in 23% of all the ALD patients. Further, 7% of the 73 non-ALD patients had a positive hEtG result (Table 2). GGT was more often elevated among hEtG-positive patients than among hEtG-negative patients, indicating the harmful effect of alcohol on graft function. But the PPV of an elevated GGT for detecting patients with chronic alcohol consumption was only 20%.
CONCLUSIONS In conclusion, hEtG determination is a valuable tool for revealing regular alcohol consumption among liver transplant recipients. Patients’ self-reports underestimate the amount of alcohol consumed and physicians’ assessments were also not very reliable. To identify patients early who may benefit from professional alcohol counseling, we now have implemented hEtG testing in the annual routine check-up of all liver transplant recipients with ALD and of all non-ALD patients who are suspected of alcohol consumption by their physician. REFERENCES 1. Pfitzmann R, Schwenzer J, Rayes N, et al. Long-term survival and predictors of relapse after orthotopic liver transplantation for alcoholic liver disease. Liver Transpl. 2007;13:197–205.
Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Ther Drug Monit Volume 37, Number 4, August 2015
2. DiMartini A, Day N, Dew M, et al. Alcohol consumption patterns and predictors of use following liver transplantation for alcoholic liver disease. Liver Transpl. 2006;12:813–820. 3. Cuadrado A, Fábrega E, Casafont F, et al. Alcohol recidivism impairs long-term patient survival after orthotopic liver transplantation for alcoholic liver disease. Liver Transpl. 2005;11:420–426. 4. DiMartini A, Dew M. Monitoring alcohol use on the liver transplant wait list: therapeutic and practical issues. Liver Transpl. 2012;18:1267–1269. 5. Stibler H. Carbohydrate-deficient transferrin in serum: a new marker of potentially harmful alcohol consumption reviewed. Clin Chem. 1991;37: 2029–2037. 6. Helander A, Eriksson C. Laboratory tests for acute alcohol consumption: results of the WHO/ISBRA Study on State and Trait Markers of Alcohol Use and Dependence. Alcohol Clin Exp Res. 2002;26:1070–1077. 7. Thierauf A, Halter CC, Rana S, et al. Urine tested positive for ethyl glucuronide after trace amounts of ethanol. Addiction. 2009;104:2007– 2012. 8. Wurst FM, Kempter C, Seidl S, et al. Ethyl glucuronide—a marker of alcohol consumption and a relapse marker with clinical and forensic implications. Alcohol Alcohol. 1999;34:71–77. 9. Staufer K, Andresen H, Vettorazzi E, et al. Urinary ethyl glucuronide as a novel screening tool in patients pre- and post-liver transplantation improves detection of alcohol consumption. Hepatology. 2011;54: 1640–1649. 10. Piano S, Marchioro L, Gola E, et al. Assessment of alcohol consumption in liver transplant candidates and recipients: the best combination of the tools available. Liver Transpl. 2014;20:815–822. 11. SOHT. 2014 Consensus for the use of alcohol markers in hair for assessment of both abstinence and chronic excessive alcohol consumption. Available at: http://www.soht.org/images/pdf/2014%20Alcohol%20markers %20revision%2013JUN14%20FINAL.pdf. Accessed September 10, 2014. 12. Agius R, Ferreira L, Yegles M. Can ethyl glucuronide in hair be determined only in 3 cm hair strands? Forensic Sci Int. 2012;218:3–9. 13. Appenzeller BM, Agirman R, Neuberg P, et al. Segmental determination of ethyl glucuronide in hair: a pilot study. Forensic Sci Int. 2007; 173:87–92. 14. Politi L, Morini L, Leone F, et al. Ethyl glucuronide in hair: is it a reliable marker of chronic high levels of alcohol consumption? Addiction. 2006; 101:1408–1412. 15. Suesse S, Pragst F, Mieczkowski T, et al. Practical experiences in application of hair fatty acid ethyl esters and ethyl glucuronide for detection of chronic alcohol abuse in forensic cases. Forensic Sci Int. 2012;218:82–91. 16. Paul L, Mußhoff F, Aebi B, et al. Richtlinie der GTFCh zur Qualitätssicherung bei forensisch-toxikologischen Untersuchungen (guidelines for quality assurance in forensic-toxicological analyses) [in German]. Available at: http://wwwgtfchorg/cms/images/stories/files/GTFCh_Richtlinie_ For-Tox_Version%201pdf. Accessed September 6, 2009. 17. Huckenbeck W, Bonte W. Alkohologie. In: Madea B, Brinkmann B, eds. Handbuch Gerichtliche Medizin. Heidelberg, Germany: Springer; 2003. 18. Hempel J, Greif-Higer G, Kaufmann T, et al. Detection of alcohol consumption in patients with alcoholic liver cirrhosis during the
Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
Hair EtG in Patients After Liver Transplantation
19.
20. 21.
22.
23.
24. 25. 26. 27. 28. 29. 30. 31. 32. 33.
evaluation process for liver transplantation. Liver Transpl. 2012;18: 1310–1315. Kharbouche H, Sporkert F, Troxler S, et al. Development and validation of a gas chromatography-negative chemical ionization tandem mass spectrometry method for the determination of ethyl glucuronide in hair and its application to forensic toxicology. J Chromatogr B Analyt Technol Biomed Life Sci. 2009;877:2337–2343. Lees R, Kingston R, Williams T, et al. Comparison of ethyl glucuronide in hair with self-reported alcohol consumption. Alcohol Alcohol. 2012; 47:267–272. Hastedt M, Büchner M, Rothe M, et al. Detecting alcohol abuse: traditional blood alcohol markers compared to ethyl glucuronide (EtG) and fatty acid ethyl esters (FAEEs) measurement in hair. Forensic Sci Med Pathol. 2013;9:184–193. Boscolo-Berto R, Favretto D, Cecchetto G, et al. Sensitivity and specificity of EtG in hair as a marker of chronic excessive drinking: pooled analysis of raw data and meta-analysis of diagnostic accuracy studies. Ther Drug Monit 2014;36:560–567. Sterneck M, Yegles M, von Rothkirch G, et al. Determination of ethyl glucuronide in hair improves evaluation of long-term alcohol abstention in liver transplant candidates. Liver Int. 2013. doi: 10.1111/liv.12243. Epub ahead of print. Stewart S, Koch D, Willner I, et al. Hair ethyl glucuronide is highly sensitive and specific for detecting moderate-to-heavy drinking in patients with liver disease. Alcohol Alcohol. 2013;48:83–87. Gareri J, Appenzeller B, Walasek P, et al. Impact of hair-care products on FAEE hair concentrations in substance abuse monitoring. Anal Bioanal Chem. 2011;400:183–188. Martins Ferreira L, Binz T, Yegles M. The influence of ethanol containing cosmetics on ethyl glucuronide concentration in hair. Forensic Sci Int. 2012;218:123–125. Yegles M, Labarthe A, Auwarter V, et al. Comparison of ethyl glucuronide and fatty acid ethyl ester concentrations in hair of alcoholics, social drinkers and teetotallers. Forensic Sci Int. 2004;145:167–173. Morini L, Politi L, Polettini A. Ethyl glucuronide in hair. A sensitive and specific marker of chronic heavy drinking. Addiction. 2009;104:915–920. Crunelle C, Cappelle D, Covaci A, et al. Hair ethyl glucuronide as a biomarker of alcohol consumption in alcohol-dependent patients: role of gender differences. Drug Alcohol Depend. 2014;141:163–166. Sporkert F, Kharbouche H, Augsburger MP, et al. Positive EtG findings in hair as a result of a cosmetic treatment. Forensic Sci Int. 2012;218: 97–100. Kintz P, Nicholson D. Testing for ethanol markers in hair: discrepancies after simultaneous quantification of ethyl glucuronide and fatty acid ethyl esters. Forensic Sci Int. 2014;13:44–46. Kintz P, Nicholson D. Interpretation of a highly positive ethyl glucuronide together with negative fatty acid ethyl esters in hair and negative blood results. Forensic Toxicol. 2014;32:176–179. Høiseth G, Morini L, Ganss R, et al. Higher levels of hair ethyl glucuronide in patients with decreased kidney function. Alcohol Clin Exp Res. 2013;37(suppl 1):E14–E16.
545
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
