- Split View
-
Views
-
Cite
Cite
Jan Toralf Fosen, Luca Morini, Cristina Sempio, Nefele Giarratana, Asle Enger, Jørg Mørland, Gudrun Høiseth, Ethyl Glucuronide Elimination Kinetics in Fingernails and Comparison to Levels in Hair, Alcohol and Alcoholism, Volume 52, Issue 5, September 2017, Pages 580–586, https://doi.org/10.1093/alcalc/agx035
- Share Icon Share
Abstract
Measurement of ethyl glucuronide (EtG) in nail, as a biomarker for alcohol intake, has recently been suggested as alternative to measurement in hair. The aim of this study was to compare levels of EtG in nail and hair, and to investigate the elimination kinetics of EtG in fingernails during an alcohol abstinent period.
Overall, 40 subjects (median estimated daily intake of ethanol (EDI) 92.5 g/day) were recruited from an alcohol rehabilitation clinic. Nail and hair samples were collected at inclusion and nail clippings were collected every 7–10th day for up to 12 weeks.
All patients showed higher nail EtG/EDI ratios compared to hair EtG/EDI ratios (P < 0.001). The median value of the ratios between EtG in nail and EtG in hair was 5.0 (range: 1.07–56.1). There was a significant correlation between nail EtG/EDI and hair EtG/EDI (Spearman's ρ = 0.638, P < 0.001). EtG disappeared from nails after ~2 months of abstinence and the median calculated EtG half-life in nail clippings was 13.3 days (range: 5.5–29.0). There was a significant correlation between the time elapsed to last positive sample for nail EtG and nail EtG levels at time of inclusion (Spearman's ρ = 0.449, P = 0.004).
The present data indicate that EtG cut-off levels in nails should be higher compared to the established 30 pg/mg EtG cut-off in hair representing heavy drinking. EtG may disappear faster from nail than expected from nail growth physiology.
Nails are an alternative matrix to hair when measuring ethyl glucuronide (EtG). The present study indicate that EtG cut-off levels in nails should be higher compared to the established 30 pg/mg EtG cut-off in hair representing heavy drinking, and EtG may disappear faster from nail than expected.
INTRODUCTION
Ethyl glucuronide (EtG), which is a minor ethanol metabolite formed by the enzyme 5′-diphospho-glucuronosyl transferase is widely used as a biomarker for alcohol intake. Traditional biomatrices like blood, plasma/serum and urine allow only limited time windows for detecting previous alcohol intake. During the recent years EtG in hair has been established as a marker of chronic alcohol consumption (Crunelle et al., 2014; Kummer et al., 2016) and is widely used in forensic toxicology. European Workplace Drug Testing Society have also included the use of EtG in hair in their guidelines (Agius and Kintz, 2010) and 30 pg/mg is used as a cut-off for heavy drinking (Kintz, 2015).
An obvious advantage for using hair and other keratin samples is the convenient and non-invasive collection process. Analyzing EtG in nails has during the last years proven to be promising as an alternative matrix to hair (Baumgartner, 2014). Several studies have investigated xenobiotics in nails (Cappelle et al., 2014; Shu et al., 2015), however, only very few studies have been published with levels of EtG in nail compared to levels of EtG in hair. Morini et al. (2012) reported higher concentrations of EtG in nail compared to hair and concluded that EtG in nail in contrast to hair may be a potential marker for binge drinking in addition to being a marker for chronic alcohol abuse. Another study reported that EtG levels in fingernail were three times higher than EtG levels in hair (Jones et al., 2012). Berger et al. found high sensitivity and specificity for EtG in both fingernails and hair as long-term alcohol markers, but quantitative individual results from nail and hair were not reported. They concluded that especially fingernails have a potential as a quantitative indicator for alcohol use, and proposed a possible cut-off of 56 pg/mg for high-risk drinkers (Berger et al., 2014).
The continuous growing of nails is an important difference to that of hair which follow a growth cycle with differences in growth speed (Baumgartner, 2014), and the mechanisms involved in nail growth are not fully understood. This complicates the understanding of incorporation of substances into nails. It is proposed that in addition to the horizontal formation of the nail in the nail root or germinal matrix, 20% of the total nail mass comes from continuous vertical growth of the nail from the surface of the nail bed; increasing the thickness of the nail (Garside, 2008). Previous literature reported fingernail growth ranging from 1.9 to 4.4 mm per month giving a potential time window for detecting drugs for ~3–5 months after cessation of use (Garside, 2008). However, the drug level in nails could be affected by contribution from more recent blood concentrations through vertical nail growth and by contamination from biological fluids (Cappelle et al., 2014; Madry et al., 2014). Nail drug levels after cessation of drug use could be influenced by drug extraction into blood circulating through the nail bed and to the exterior (by hand washing). All these processes would influence real-life detection time windows.
Regarding longitudinal data for drug levels in nails, and their relation to hair levels, only very few studies have been performed (Ropero-Miller et al., 2000; Lin et al., 2004; Hang et al., 2013; Madry et al., 2014). To the best of our knowledge, no studies have investigated EtG kinetics in fingernails after cessation of drinking, and consequently important information needed to interpret EtG results from nail samples is lacking.
The aim of this study was to compare levels of EtG in fingernails to the levels in hair of alcohol dependent subjects at the time of inclusion in a treatment program. In addition, we wanted to investigate the kinetics of EtG in fingernails of these patients during an alcohol abstinent period by taking consecutive nail samples for up to 12 weeks after end of drinking.
MATERIALS AND METHODS
Study design and protocol
Patients were recruited from a 12-step treatment clinic; Trasoppklinikken in Oslo, Norway. The clinic is an open unit staffed with specialized medical and nursing personnel. The clinical board at Trasoppklinikken approved the study. All participants signed informed consent before attending the study which was approved by the National Committee for Research Ethics in Norway (ref. no. 2013/1222).
The inpatient treatment program at Trasoppklinikken usually rounds 10 weeks, ranging from detoxification to completing educative, cognitive behavioral and 12-step therapy. The treatment program and clinic routines were not changed by the present study and all patients are referred for treatment on a voluntary basis. Most patients use only alcohol, but there are no restrictions concerning other drug use/dependencies. On admission, all patient belongings are searched for drugs. The patients are allowed to leave the clinic in groups of three for shorter errands twice per week, and breath ethanol is measured after return. Breath ethanol is also measured on suspicion of relapse. Positive breath ethanol results were recorded as a part of the present study.
The patients received a preliminary informal invitation to the study as outpatients, before admission to the clinic. After admission to the inpatient program, all patients were formally invited to join the study and signed the informed consent.
The inclusion criterion for the present study was a history of alcohol ingestion within the last three weeks before the first sample collection. Forty patients were included in the study. All patients met the ICD-10 criteria for alcohol dependence (WHO, 2010). All patients were interviewed about alcohol consumption using a modified AUDIT-test (NIDA, 2016), which included questions about frequency of ingestion combined with amount and type of ethanol consumed. An estimated daily intake of ethanol (EDI) was calculated on basis of the information obtained (representing an average consumption over the last 90 days). In the case of some abstinence days during this period, the reported EDI therefore appears lower than the actual EDI for the drinking period. Time of last alcohol intake was given special attention during the interview, and this date was recorded. In the modified AUDIT-interview, information about washing/treatment of nails and hair was obtained. Only three patients reported using nail polish during the last 3 months. In total, 39 patients reported their hand washing routines; all these washed their hands at least twice a day. Overall, 13 patients reported hair bleaching, coloring or perming for the last 3 months. Of these, treatment was specified as coloring in five patients and hair bleaching in one patient.
In addition to the nail sample collection at time of inclusion a hair sample was also provided for comparison. The hair was collected by cutting a hair specimen as close as possible to the skin at the posterior vertex. The 3 cm proximal hair segment, roughly representing growth the last 3 months, was submitted for EtG analysis. EtG-test results were neither disclosed to the patients nor to the treatment staff. Other laboratory parameters were measured relatively close to admission and were received from the medical record. These included serum creatinine, aspartate aminotransferase (AST), alanine aminotransferase (ALT), gamma-glutamyl transferase (GGT) and carbohydrate deficient transferrin (CDT). Values above the normal range (Schellenberg and Wielders, 2010; Fürst Medical Laboratory, 2016) were seen for 11 patients for CDT% (known for 65 % of the included patients), 17 patients for AST, 15 patients for ALT, 16 patients for GGT and 1 patient for creatinine. In addition height, weight, age and sex were recorded for each participant. All subjects were inpatients during the total study period. Nail clippings were collected approximately once every 7–10th day as long as the patient remained in the clinic by cutting the nail overhang from the thumbs and the index fingers on both hands. All nail samples were analyzed for EtG. It was also noted if the clinic staff suspected alcohol intake since collection of the last sample. This could be based on a result of breath ethanol analysis or for other clinically valid reasons.
Analytical methods
Both methods used for analyzing nails and hair have been fully validated and are previously published (Morini et al., 2006, 2012). Level of detection (LOD) and limit of quantification (LOQ) were 6.0 and 10.0 pg/mg for nail and 3.0 and 5.0 pg/mg for hair, respectively.
Chemicals and materials
EtG and D5-EtG were obtained from Medichem (Promochem, Milan, Italy). Water was purified by filtering deionized water on a Milli-Q Simplicity 185 filtration system from Millipore (Bedford, MA, USA). Formic acid for mass spectrometry was purchased from Sigma-Aldrich (St Louis, MI, USA). HPLC-grade methanol and acetonitrile were obtained from Panreac (NovaChimica, Milan, Italy).
Sample preparation
Approximately 10–20 mg nails per sample were submitted to a 2-step washing procedure using dichloromethane and methanol. After complete removal of methanol, samples were dried under nitrogen stream. Samples were then soaked in 300 μL of water, and 20 μL of internal standard (D5-EtG) aqueous solution (0.01 mg/L), and sonicated up to 2 h. Finally, 10 μL of supernatant was injected in the LC–MS/MS system.
Approximately 50 mg hair samples were washed with organic solvents (dichloromethane and methanol). They were dried and cut into small pieces (1–2 mm length). Samples were then soaked in 700 μL of water, and 20 μL of internal standard (D5-EtG) aqueous solution (0.01 mg/L), and sonicated up to 2 h. Finally, 8 μL of supernatant was injected in the LC–MS/MS system.
Instrumentation
LC–MS/MS analyses were carried out through an Agilent 1100/1200 Series system (Agilent Technologies, Palo Alto, CA, USA) interfaced to a 4000 Q-TRAP (Applied Biosystem/MDS SCIEX, Foster City, CA, USA) with an electrospray (ESI) Turbo V™ Ion Source. The Kinetex C18 coulmn (100 × 2.1 mm2 i.d., 5-μm particle size) was kept at 25°C during the analysis. Analytes were eluted in isocratic mode, using a mobile phase consisting of 99% formic acid (0.1%) and 1% acetonitrile. A post-column elution of 0.1 mL/min of acetonitrile was used to enhance sensitivity.
The ESI source settings were: ion-spray voltage, 4200 V; source temperature, 400°C; nebulization and heating gas (air), 40 and 50, respectively. Multiple reaction monitoring was optimized using nitrogen as collision gas (with pressure set at level 7) and a dwell time of 100 ms.
The transitions m/z 221 → 75 and m/z 226 → 75 were used as quantifiers for EtG and D5-EtG, respectively. Data acquisition and elaboration were provided by Analyst® software (version 1.5.1 AB SCIEX).
Statistics
Mean and standard deviation (SD) for continuous variables and frequency distribution for categorical variables are reported for patient characteristics. The time of inclusion (upon admission to the clinic when the first nail sample was collected) was set as time zero and intervals between samples reported as days after the first sample. Concentrations of EtG in nail and hair were not normally distributed. For all results, median and range values are reported for continuous variables and frequency distribution is reported for categorical variables.
Correlation between continuous variables was assessed using the non-parametric Spearman's rank correlation test. Differences between dependent samples like concentrations in nail and hair within each subject were studied using the non-parametric Wilcoxon signed ranks test. Differences between independent samples were analyzed using the non-parametric Mann–Whitney U-test.
The Kinetica software version 5.1 (Thermo Fisher Scientific Inc., Waltham, MA, USA) was used for computing nail EtG half-life. The software's standard parameters were used.
RESULTS
A total of 40 patients (62.5% men) with an average age of 47.0 years (SD: 10.6) were included in the study. Each patient was included for an average of 51.8 (SD: 25.8) days. The minimum number of days included in the study was 1 (n = 2), and the highest number of days was 91 (n = 1). Nail samples were collected approximately once a week giving a mean total of 6.1 (SD: 2.9) nail samples per patient. The minimum number of collected nail samples per patient was 1 (n = 2), and the maximum number per patient was 11 (n = 6). In total, 243 nail samples were collected. One hair sample was collected from each patient at inclusion, but for two of the patients there was not enough hair material for analysis. Hair results are therefore based on 38 patients.
Based on self-report of average alcohol intake during the last 3 months, the median EDI was 92.5 g/day (range: 8.0– 455.0). A median time of 8.5 days (range: 2–18) had passed since the last alcohol intake and inclusion in the present study. Intake of alcohol during the study period was suspected for three patients. Individual values of nail- and hair EtG at inclusion related to EDI (and last intake of alcohol) are shown in Table 1. The median level of EtG was 252.7 pg/mg (range: 27.1–2267) for the nail samples taken at inclusion. For the hair samples collected simultaneously, the median level of EtG was 49.1 pg/mg (range: 0–718.5). In one patient, the EtG concentration in hair was below LOQ (this patient reported an EDI of 75 g/day, but had been abstinent for the last 7 days before inclusion). The relationships between individual nail EtG and hair EtG levels at inclusion and reported EDI are shown in Fig. 1a and b.
Patient number . | Abstinent period before inclusion (days) . | EDIa (g/day) . | Nail EtG (pg/mg) . | Hair EtG (pg/mg) . | Last positive sample (days after inclusion) . |
---|---|---|---|---|---|
1 | 4 | 8.0 | 254.0 | 12.0 | 49b |
2 | 14 | 9.5 | 114.0 | 50.5 | 26 |
3 | 6 | 15.0 | 298.5 | – | 37b |
4 | 9 | 15.1 | 53.8 | 11.8 | 12b |
5 | 6 | 16.0 | 110.7 | – | 7 |
6 | 5 | 17.0 | 233.3 | 31.1 | 46 |
7 | 8 | 27.5 | 55.2 | 8.0 | 61b |
8 | 13 | 28.0 | 225.0 | 59.6 | 32b |
9 | 7 | 45.0 | 27.1 | 12.4 | 17b |
10 | 14 | 50.0 | 120.0 | 17.4 | 28 |
11 | 2 | 50.0 | 829.0 | 40.4 | 34 |
12 | 6 | 51.0 | 187.5 | 14.2 | 41 |
13 | 12 | 51.0 | 120.0 | 100.0 | 29 |
14 | 7 | 57.0 | 825.0 | 54.5 | 41 |
15 | 10 | 60.0 | 63.0 | 16.0 | 27 |
16 | 18 | 66.0 | 2177.8 | 38.8 | 43 |
17 | 13 | 71.0 | 113.2 | 47.5 | 41 |
18 | 9 | 72.0 | 2266.7 | 179.8 | 61b |
19 | 7 | 75.0 | 52.7 | <LOQ | 12b |
20 | 11 | 85.0 | 122.9 | 7.7 | 28 |
21 | 5 | 100.0 | 43.1 | 29.6 | 17b |
22 | 11 | 100.0 | 337.2 | 48.2 | 81b |
23 | 4 | 120.0 | 121.4 | 82.0 | 0b |
24 | 4 | 123.0 | 129.0 | 78.0 | 54b |
25 | 12 | 144.0 | 560.0 | 111.9 | 47 |
26 | 11 | 150.0 | 508.3 | 180.2 | 10 |
27 | 5 | 150.0 | 428.3 | 94.5 | 61b |
28 | 9 | 156.0 | 163.0 | 58.9 | 33b |
29 | 3 | 160.0 | 655.0 | 279.0 | 39 |
30 | 9 | 162.0 | 66.3 | 13.0 | 26 |
31 | 9 | 170.0 | 257.4 | 23.0 | 0 |
32 | 2 | 180.0 | 1194.4 | 226.7 | 55 |
33 | 3 | 200.0 | 283.0 | 265.0 | 42b |
34 | 11 | 225.0 | 729.0 | 401.2 | 19b |
35 | 2 | 280.0 | 285.7 | 102.0 | 0b |
36 | 14 | 300.0 | 634.4 | 50.0 | 40 |
37 | 2 | 300.0 | 1552.5 | 60.0 | 36 |
38 | 14 | 450.0 | 832.6 | 718.5 | 40 |
39 | 2 | 450.0 | 371.0 | 47.5 | 36b |
40 | 12 | 455.0 | 251.4 | 46.0 | 32 |
Patient number . | Abstinent period before inclusion (days) . | EDIa (g/day) . | Nail EtG (pg/mg) . | Hair EtG (pg/mg) . | Last positive sample (days after inclusion) . |
---|---|---|---|---|---|
1 | 4 | 8.0 | 254.0 | 12.0 | 49b |
2 | 14 | 9.5 | 114.0 | 50.5 | 26 |
3 | 6 | 15.0 | 298.5 | – | 37b |
4 | 9 | 15.1 | 53.8 | 11.8 | 12b |
5 | 6 | 16.0 | 110.7 | – | 7 |
6 | 5 | 17.0 | 233.3 | 31.1 | 46 |
7 | 8 | 27.5 | 55.2 | 8.0 | 61b |
8 | 13 | 28.0 | 225.0 | 59.6 | 32b |
9 | 7 | 45.0 | 27.1 | 12.4 | 17b |
10 | 14 | 50.0 | 120.0 | 17.4 | 28 |
11 | 2 | 50.0 | 829.0 | 40.4 | 34 |
12 | 6 | 51.0 | 187.5 | 14.2 | 41 |
13 | 12 | 51.0 | 120.0 | 100.0 | 29 |
14 | 7 | 57.0 | 825.0 | 54.5 | 41 |
15 | 10 | 60.0 | 63.0 | 16.0 | 27 |
16 | 18 | 66.0 | 2177.8 | 38.8 | 43 |
17 | 13 | 71.0 | 113.2 | 47.5 | 41 |
18 | 9 | 72.0 | 2266.7 | 179.8 | 61b |
19 | 7 | 75.0 | 52.7 | <LOQ | 12b |
20 | 11 | 85.0 | 122.9 | 7.7 | 28 |
21 | 5 | 100.0 | 43.1 | 29.6 | 17b |
22 | 11 | 100.0 | 337.2 | 48.2 | 81b |
23 | 4 | 120.0 | 121.4 | 82.0 | 0b |
24 | 4 | 123.0 | 129.0 | 78.0 | 54b |
25 | 12 | 144.0 | 560.0 | 111.9 | 47 |
26 | 11 | 150.0 | 508.3 | 180.2 | 10 |
27 | 5 | 150.0 | 428.3 | 94.5 | 61b |
28 | 9 | 156.0 | 163.0 | 58.9 | 33b |
29 | 3 | 160.0 | 655.0 | 279.0 | 39 |
30 | 9 | 162.0 | 66.3 | 13.0 | 26 |
31 | 9 | 170.0 | 257.4 | 23.0 | 0 |
32 | 2 | 180.0 | 1194.4 | 226.7 | 55 |
33 | 3 | 200.0 | 283.0 | 265.0 | 42b |
34 | 11 | 225.0 | 729.0 | 401.2 | 19b |
35 | 2 | 280.0 | 285.7 | 102.0 | 0b |
36 | 14 | 300.0 | 634.4 | 50.0 | 40 |
37 | 2 | 300.0 | 1552.5 | 60.0 | 36 |
38 | 14 | 450.0 | 832.6 | 718.5 | 40 |
39 | 2 | 450.0 | 371.0 | 47.5 | 36b |
40 | 12 | 455.0 | 251.4 | 46.0 | 32 |
aAverage over the past 90 days.
bFor these patients, last positive sample is set to last sample, as negative samples were not reached during the study period.
Patient number . | Abstinent period before inclusion (days) . | EDIa (g/day) . | Nail EtG (pg/mg) . | Hair EtG (pg/mg) . | Last positive sample (days after inclusion) . |
---|---|---|---|---|---|
1 | 4 | 8.0 | 254.0 | 12.0 | 49b |
2 | 14 | 9.5 | 114.0 | 50.5 | 26 |
3 | 6 | 15.0 | 298.5 | – | 37b |
4 | 9 | 15.1 | 53.8 | 11.8 | 12b |
5 | 6 | 16.0 | 110.7 | – | 7 |
6 | 5 | 17.0 | 233.3 | 31.1 | 46 |
7 | 8 | 27.5 | 55.2 | 8.0 | 61b |
8 | 13 | 28.0 | 225.0 | 59.6 | 32b |
9 | 7 | 45.0 | 27.1 | 12.4 | 17b |
10 | 14 | 50.0 | 120.0 | 17.4 | 28 |
11 | 2 | 50.0 | 829.0 | 40.4 | 34 |
12 | 6 | 51.0 | 187.5 | 14.2 | 41 |
13 | 12 | 51.0 | 120.0 | 100.0 | 29 |
14 | 7 | 57.0 | 825.0 | 54.5 | 41 |
15 | 10 | 60.0 | 63.0 | 16.0 | 27 |
16 | 18 | 66.0 | 2177.8 | 38.8 | 43 |
17 | 13 | 71.0 | 113.2 | 47.5 | 41 |
18 | 9 | 72.0 | 2266.7 | 179.8 | 61b |
19 | 7 | 75.0 | 52.7 | <LOQ | 12b |
20 | 11 | 85.0 | 122.9 | 7.7 | 28 |
21 | 5 | 100.0 | 43.1 | 29.6 | 17b |
22 | 11 | 100.0 | 337.2 | 48.2 | 81b |
23 | 4 | 120.0 | 121.4 | 82.0 | 0b |
24 | 4 | 123.0 | 129.0 | 78.0 | 54b |
25 | 12 | 144.0 | 560.0 | 111.9 | 47 |
26 | 11 | 150.0 | 508.3 | 180.2 | 10 |
27 | 5 | 150.0 | 428.3 | 94.5 | 61b |
28 | 9 | 156.0 | 163.0 | 58.9 | 33b |
29 | 3 | 160.0 | 655.0 | 279.0 | 39 |
30 | 9 | 162.0 | 66.3 | 13.0 | 26 |
31 | 9 | 170.0 | 257.4 | 23.0 | 0 |
32 | 2 | 180.0 | 1194.4 | 226.7 | 55 |
33 | 3 | 200.0 | 283.0 | 265.0 | 42b |
34 | 11 | 225.0 | 729.0 | 401.2 | 19b |
35 | 2 | 280.0 | 285.7 | 102.0 | 0b |
36 | 14 | 300.0 | 634.4 | 50.0 | 40 |
37 | 2 | 300.0 | 1552.5 | 60.0 | 36 |
38 | 14 | 450.0 | 832.6 | 718.5 | 40 |
39 | 2 | 450.0 | 371.0 | 47.5 | 36b |
40 | 12 | 455.0 | 251.4 | 46.0 | 32 |
Patient number . | Abstinent period before inclusion (days) . | EDIa (g/day) . | Nail EtG (pg/mg) . | Hair EtG (pg/mg) . | Last positive sample (days after inclusion) . |
---|---|---|---|---|---|
1 | 4 | 8.0 | 254.0 | 12.0 | 49b |
2 | 14 | 9.5 | 114.0 | 50.5 | 26 |
3 | 6 | 15.0 | 298.5 | – | 37b |
4 | 9 | 15.1 | 53.8 | 11.8 | 12b |
5 | 6 | 16.0 | 110.7 | – | 7 |
6 | 5 | 17.0 | 233.3 | 31.1 | 46 |
7 | 8 | 27.5 | 55.2 | 8.0 | 61b |
8 | 13 | 28.0 | 225.0 | 59.6 | 32b |
9 | 7 | 45.0 | 27.1 | 12.4 | 17b |
10 | 14 | 50.0 | 120.0 | 17.4 | 28 |
11 | 2 | 50.0 | 829.0 | 40.4 | 34 |
12 | 6 | 51.0 | 187.5 | 14.2 | 41 |
13 | 12 | 51.0 | 120.0 | 100.0 | 29 |
14 | 7 | 57.0 | 825.0 | 54.5 | 41 |
15 | 10 | 60.0 | 63.0 | 16.0 | 27 |
16 | 18 | 66.0 | 2177.8 | 38.8 | 43 |
17 | 13 | 71.0 | 113.2 | 47.5 | 41 |
18 | 9 | 72.0 | 2266.7 | 179.8 | 61b |
19 | 7 | 75.0 | 52.7 | <LOQ | 12b |
20 | 11 | 85.0 | 122.9 | 7.7 | 28 |
21 | 5 | 100.0 | 43.1 | 29.6 | 17b |
22 | 11 | 100.0 | 337.2 | 48.2 | 81b |
23 | 4 | 120.0 | 121.4 | 82.0 | 0b |
24 | 4 | 123.0 | 129.0 | 78.0 | 54b |
25 | 12 | 144.0 | 560.0 | 111.9 | 47 |
26 | 11 | 150.0 | 508.3 | 180.2 | 10 |
27 | 5 | 150.0 | 428.3 | 94.5 | 61b |
28 | 9 | 156.0 | 163.0 | 58.9 | 33b |
29 | 3 | 160.0 | 655.0 | 279.0 | 39 |
30 | 9 | 162.0 | 66.3 | 13.0 | 26 |
31 | 9 | 170.0 | 257.4 | 23.0 | 0 |
32 | 2 | 180.0 | 1194.4 | 226.7 | 55 |
33 | 3 | 200.0 | 283.0 | 265.0 | 42b |
34 | 11 | 225.0 | 729.0 | 401.2 | 19b |
35 | 2 | 280.0 | 285.7 | 102.0 | 0b |
36 | 14 | 300.0 | 634.4 | 50.0 | 40 |
37 | 2 | 300.0 | 1552.5 | 60.0 | 36 |
38 | 14 | 450.0 | 832.6 | 718.5 | 40 |
39 | 2 | 450.0 | 371.0 | 47.5 | 36b |
40 | 12 | 455.0 | 251.4 | 46.0 | 32 |
aAverage over the past 90 days.
bFor these patients, last positive sample is set to last sample, as negative samples were not reached during the study period.
There was a significant correlation between the level of EtG in nail and EDI (Spearman's ρ = 0.440, P = 0.004). There was also a significant correlation between hair EtG and EDI (Spearman's ρ = 0.519, P = 0.001). There was no significant correlation between AST, ALT or CDT and EDI, but a significant correlation between GGT and EDI (Spearman's ρ = 0.385, P < 0.019).
The nail EtG/EDI-ratio and hair EtG/EDI-ratio are shown in Fig. 2. The nail EtG/EDI-ratio ranged between 0.41 and 33.0 (pg/mg)/(g/d) (mean: 6.3, median: 2.6). The hair EtG/EDI-ratio (when excluding the single patient negative for EtG in hair) ranged between 0.08 and 5.3 (pg/mg)/(g/d) (mean: 0.9, median: 0.6).
The median value of the ratios between EtG concentrations in nail and hair was 5.0 (range: 1.07–56.1). There was a significant correlation between nail EtG/EDI and hair EtG/EDI (Spearman's ρ = 0.638, P < 0.001), as well as the concentrations of EtG in nail and hair (not corrected for EDI) (Spearman's ρ = 0.631, P < 0.001).
Individual longitudinal data for EtG values in nail for all patients are shown in Fig. 3. In total, 30 patients showed maximum nail EtG concentrations at the time of inclusion. Twenty-seven of these showed decreasing (or within an analytical variation of ~20%) values during the total study period, while three patients showed an increase in concentration during the study period. In two of these three patients, an alcohol intake was suspected during the period before the samples showing increasing concentrations. One of the two patients showing increasing EtG levels after suspected relapse had EtG level below LOQ before the suspected alcohol intake and reached an EtG concentration of 105.4 pg/mg 10 days after the suspected intake.
In another eight of the patients, maximum levels were reached at a later stage than inclusion, median 12.0 days (range: 7–28) after inclusion. For these eight patients, there was no suspicion of intake shortly after inclusion which could explain this finding. From the remaining two patients only one nail sample was available.
A total of 22 patients reached negative EtG nail samples during the study period. For these, the median time to achieve negative samples was 47.0 days (range: 13–60). Among all 40 patients, the last positive EtG in nail (before the concentration was below LOQ) was seen after median 36 days (range: 0–81). For those not reaching levels below LOQ during the study period, the last sample was registered as the last positive sample.
There was a significant correlation between the time elapsed till the last positive sample and nail EtG levels at time of inclusion (Spearman's ρ = 0.449, P = 0.004). There was no significant correlation between time for last positive sample and reported EDI (Spearman's ρ = −0.017, P = 0.918).
In 28 patients there was enough consecutive samples showing decreasing values (or within analytical variation) for half-life calculations using the Kinetica software. In these, the median calculated half-life for EtG in nail was 13.3 days (range: 5.5–29.0).
DISCUSSION
The main purpose of this study was to compare nail EtG levels to hair EtG levels, and to investigate the kinetics of EtG in fingernails during an alcohol abstinent period. The present study showed that EtG levels are higher in nail compared to hair. We also presented kinetic parameters for EtG in nail by taking consecutive nail samples after end of drinking.
At inclusion, EtG levels in nail were significantly higher than in hair for all subjects, supporting the results obtained previously (Jones et al., 2012; Berger et al., 2014), while the opposite situation is previously published for zolpidem, codeine, cocaine and MDMA in fingernail clippings or scrapings (Ropero-Miller et al., 2000; Lin et al., 2004; Hang et al., 2013; Madry et al., 2016). Nowadays, EtG in hair has been accepted as a reliable marker for chronic excessive alcohol consumption diagnosis (Kintz, 2015). It is previously suggested that the heavy drinking EtG cut-off should be higher in nails compared to hair (Berger et al., 2014).While Berger et al. (2014) proposed 56 pg/mg as possible threshold for high-risk drinkers, our results suggest a more precautionary approach. The number of subjects in the present study is too limited to provide any adequate statistical approach, but the ratio between the individual concentrations in nail and hair indicates that the nail EtG levels that corresponds to 30 pg/mg in hair (cut-off for heavy drinking) could be higher for nails.
Another outcome of the present study was the demonstration of a good correlation between levels of EtG in nails and EDI, a fact that strengthens the value of nail as an alternative matrix to hair. A comparable relation between hair EtG levels and EDI is previously highlighted in the literature (Hoiseth et al., 2009; Morini et al., 2009).
The good correlation between the last positive nail sample and the EtG level at the time of inclusion indicates that the higher concentration leads to longer window of detection. Fingernail growth implies a potential time window for detecting drugs for ~3–5 months limited by the nail growth speed. Interestingly, the present study showed that EtG in nails in most cases disappeared within ~2 months, not comparable to nail growth. Therefore, it is reasonable to state that this molecule has been removed from the keratin matrix. Although the calculation of half-lives based on nail concentrations could be subject to uncertainties, and the range of half-lives demonstrated a large variation, the calculated median half-life also indicates that EtG is lost faster than expected from nail growth physiology. Studying the maximum concentrations compared to the LOQ in the present study indicate that after 4–5 half-lives, EtG would be no longer detected in nail. Therefore, the median calculated half-life indicates a detection time of ~2 months.
There could be several reasons for loss of EtG. Firstly, EtG is a hydrophilic substance and nails are generally in contact with water several times a day. It has been shown that water transport into the nail is relatively fast (Chiu et al., 2015). This fact could be the main reason for this significantly loss of EtG. Secondly, EtG concentrations could decrease if nail without EtG from vertical growth mix with the old EtG containing matrix. Also, EtG could possibly be removed from the nail bed and back into the blood stream during an abstinent period with no EtG present in blood. Thirdly, the loss of EtG could partly be explained by poor incorporation into the nail, and reflecting that concentration at inclusion was affected by for instance EtG present in sweat. The calculated half-life in nail indicates that a few weeks of abstinence before sample collection would strongly affect the measured level of EtG. The fact that there is a gradual decrease in concentrations of EtG in nail, we believe strengthens the hypothesis of loss of EtG during nail growth, as an abrupt fall in concentrations would be expected if horizontal growth was the sole mechanism.
Longitudinal data on EtG in nails have been missing in the literature, but the present longitudinal results could only be compared to previous studies showing longitudinal data for other drugs. One previous study showed that all of the eight included patients had negative amphetamine/methamphetamine nail samples 12 weeks after cessation of use (Lin et al., 2004), results that were comparable to ours. On the contrary, two other previous studies found zolpidem in nail samples up to 3.5 months (Madry et al., 2014) and more than 6 months (Hang et al., 2013) after a single intake. In the last study, fingernails from all fingers were used and slower nail growth speed was suggested as an explanation for the longer window of detection than the study of Madry et al. (2014).
In the present study we have compared EtG levels in nails and hair at the time of inclusion and investigated the kinetics of EtG in nails of frequent drinkers. A weakness of this paper is the relatively low number of subjects that joined the study. Moreover, several conditions, such as amount and time of alcohol intake, food and drink ingestion and nail and hair treatments, could not be fully controlled. Analyzing EtG in nail and hair after pulverization of the matrices have been shown to increase EtG recovery from keratin matrix (Albermann et al., 2012; Monch et al., 2013; Cappelle et al., 2016). However, the established hair EtG cut-off for heavy drinking is based on methods where hair is both pulverized and not pulverized. The present study did not use pulverization technique even though the extraction of EtG could have been optimized by this procedure. Future studies on the importance of pulverization for the EtG cut-off values for heavy drinking is needed and eventual new cut-off values should be evaluated for both nail and hair accordingly to this procedure.
The strength of this study was the possibility to examine the fate of EtG in nails in frequent drinkers during a long period of time after end of drinking. This gives information of relevance for the interpretation of EtG results from single nail samples, but analysis of EtG in consecutive nail samples could also be useful in evaluation of patient's adherence to abstinence programs. Furthermore, an experimental study involving chronic excessive alcohol consumption would have been impossible to perform due to ethical reasons.
In conclusion, we found that EtG in nails show a good correlation with both EDI and EtG hair values, and that EtG in nail could be a potentially good alternative to hair matrix, although a reliable cut-off for heavy drinking is still to be determined. This study is the first to evaluate kinetics of EtG in nails of frequent alcohol consumers after cessation of drinking. Although individual differences, the calculated half-lives and the observations from the present study indicated that EtG disappeared from nails within ~2 months of abstinence, faster than would be assumed from nail growth only. Further studies on the mechanisms underlying EtG elimination kinetics in nails are, however, needed to fully evaluate the potential of nail EtG as a biomarker of ethanol intake.
ACKNOWLEDGMENTS
None.
FUNDING
No funding was obtained for the authorship of this article.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
ETHICAL APPROVAL
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.