Laboratory testing for recent alcohol consumption - CiteSeerX

Clinical 618-624

Chemist’y (1996)

42:4

Laboratory testing for recent alcohol consumption: comparison of ethanol, methanol, and 5-hydroxytryptophol ANDERS

HELANDER,l*

OLOF

BECK,2

alcohol alcohol

terms of diagnostic sensitivity and specificity, most studies still rely on self-reports as the “gold standard” for alcohol consumption, which creates a problem. The limited reliability of interview data underscores the need for more objective tests of recent alcohol consumption, permitting frequent monitoring of individual drinking patterns. The most obvious way to prove that a person has been drinking is by demonstrating the presence of ethanol in blood, urine, saliva, or breath [6]. However, because ethanol is rapidly eliminated from the body, this approach is practical only for consumption of alcohol within the past few hours. Increased concentrations of the metabolites of ethanol oxidation, acetaldehyde and acetic acid, have been proposed as alternative indicators of recent drinking. Acetaldehyde, however, is rapidly converted to acetic acid by aldehyde dehydrogenase (ALDH), and the concentration in peripheral blood is normally very low f7]4 Its use is further hampered by problems associated with artifactual formation after sampling the blood. The concentration of acetic acid, on the other hand, increases with the development of metabolic tolerance to alcohol, and might therefore be more suitable as a marker of chronic alcoholism [8]. A metabolic interaction between ethanol and serotonin (5hydroxytryptamine) can be utilized for detection of recent drinking [9-11]. 5-Hydroxytryptophol (5HTOL) is normally a minor metabolite of serotonin, but its formation increases dramatically after ingestion of ethanol, whereas 5-hydroxyindole-3-acetic acid (5HIAA), the major metabolite under normal

A major problem in the early detection and treatment of alcohol dependence is obtaining reliable information about a person’s drinking habits. Self-reports and diagnostic questionnaires are widely used for this purpose, but people with alcohol-related problems may deliberately deny or underreport the actual

Karolinska Institute, Department of Clinical Neuroscience, Psychiatry Section at St. Gorans Hospital, Stockholm, Sweden. 2 Department of Clinical Pharmacology, Karolinska Hospital, Stockholm, of Forensic

Toxicology,

University

Hospital,

Linkoping,

JoNEs3

dependent subjects or those engaged in hazardous drinking [4, 5]. However, when biochemical markers are evaluated in

mIt11s: alcoholism. 5-hydroxyindole-3-acetic acid. metabolism #{149} serotonin #{149} aldehyde dehydrogenase dehydrogenase

Sweden. ‘Department

WAYNE

amounts they consume [1, 2]. For this reason, underdiagnosis of alcohol misuse is not uncommon by these methods. Wellcontrolled population surveys, for example, have accounted for only about half of the known consumption of alcohol [3]. To help overcome this problem, a large number of biochemical and hematological tests have been developed to identify alcohol-

The ratio of 5-hydroxytryptophol to 5-hydroxyindole-3acetic acid (5HTOIJSHIAA) in urine was compared with concentrations of ethanol and methanol as a way to monitor recent alcohol consumption. During detoxification of alcohol-dependent subjects, ethanol persisted longer in urine than in breath or plasma. Blood and urinary methanol remained increased for 2-6 h after blood ethanol had returned to background concentrations, whereas 5HTOLI 5HIAA remained increased fir 6-15 h. In healthy volunteers who had ingested alcohol (range 3-98 g) the previous afternoon or evening, 87% (for men) and 59% (for women) of all drinking occasions exceeding 7 g of alcohol were identified by an increased 5HTOIJ5HIAA in the first morning urine void. This compares with 32% and 12%, respectively, identified by analysis of ethanol (>200 xmol/ L). No gender difference in the excretion pattern of SHTOIJSHLAA was seen. The results demonstrate that SHTOLJ5IJIAA provides a specific and more sensitive method to detect recent alcohol consumption than does ethanol or methanol. INDEXING

and A.

Swe-

Address correspondence to this author at: Alcohol & Drug Dependence Unit, St. Gorans Hospital, Box 12500, S-112 81 Stockholm, Sweden. Fax (+46) 8 6721994. Received October 10, 1995; accepted January 24, 1996.

Nonstandard 5-hydroxytryptophol; dehydrogenase.

618

abbreviations: ALDH, aldehyde 5HIAA, 5-hydroxyindole-3-acetic

dehydrogenase; 5HTOL, acid; and ADH, alcohol

Clinical Chemistiy

conditions, is concomitantly decreased. This metabolic shift can be monitored by analysis of blood or urine, and the excretion of 5HTOL will not normalize until several hours after blood and urinary ethanol have reached background concentrations [11]. On the basis of this observation, increased urinary concentration of 5HTOL was introduced as a sensitive biochemical marker of recent alcohol consumption [11, 12]. An abnormally high concentration of methanol in body fluids or breath has also been suggested as an indicator of recent drinking [13]. Methanol accumulates in the body during the metabolism of ethanol [14, 15], and, as for 5HTOL, the concentration remains increased for some time after ethanol is no longer detectable [16, 17]. Thus measurement of either 5HTOL or methanol will provide a higher sensitivity than analysis of ethanol alone for the detection of recent alcohol consumption. The present study was carried out to evaluate further the usefulness of 5HTOL as a marker of recent alcohol consumption, and to compare its sensitivity and specificity with those of ethanol and methanol.

Materials DETERMINATION

The

and Methods

#{248}F ETHANOL,

concentrations

of ethanol,

METHANOL,

methanol,

AND

ACETONE

and acetone

in blood

and urine samples were determined by headspace gas chromatography essentially as described elsewhere [18]. To increase sensitivity for determination of methanol, we used a salting-out procedure, adding 1.2 g of sodium chloride to 0.5 mL of specimen [19]. The limits of detection for analysis of ethanol and methanol were 200 j.molIL and 15 /.Lmol/L, respectively. The concentration of ethanol in breath was determined with a hand-held analyzer (Alcolmeter S-D2; Lion Labs., Barry, UK) and the result reported as presumed blood alcohol concentration. DETERMINATION

OF 5HTOL

AND

5HIAA

The concentration of 5HTOL in urine is expressed as a ratio to the concentration of 5HIAA, because ingestion of foods rich in serotonin (such as banana, pineapple, kiwifruit, and walnuts) can otherwise cause false-positive results [20]. 5HTOL and 5HIAA were determined with sensitive gas chromatographic-mass spectrometric and HPLC methods [10, 12, 2/]. The reference limit used to indicate an increased urinary 5HTOL/5HIAA ratio was set at 15 pmol/nmol [12, 22, 23]. CLINICAL

PROCEDURES

Male alcohol-dependent patients (ages 40-62 years) were recruited from the detoxification unit at St. G#{246}rans Hospital. Blood and urine samples (10 mL each) for determination of ethanol, methanol, and the 5HTOL/51-IIAA ratio were collected from 10 subjects with a blood alcohol concentration of 23.7-57.5 mmollL (mean 43.6) on admission to the hospital. Thereafter, the breath ethanol concentration was measured every 4-6 h, and blood and urine samples were also collected. When breath measurements indicated zero ethanol concentration, sampling of blood and urine was continued at 3-h intervals for another 12-15 h. In a previous experimental protocol [11], blood and urine

42, No. 4, 1996

619

samples were collected from healthy volunteers (social drinkers without any history of previous excessive consumption of alcohol). Ethanol (160 g/L in orange juice) was administered at 0.8 g/kg body weight to five women and five men (ages 23-39 years) after an overnight fast, and the drink was finished within 30 mm. From all subjects participating in that experiment, the plasma samples collected at the start of drinking and then at 2-h intervals for a total of 10 h, and the urine samples collected at the start and then at every 2-3 h for 22 h, were also used for the present study. All samples had since (for -30 months) been stored in l-mL portions at -80 #{176}C. The urinary 5HTOL/ 5HIAA ratio and the concentrations of plasma and urinary ethanol and of urinary methanol (which was not analyzed in the previous study) were determined as described above. In a new experiment on healthy volunteers, consecutive urine samples were obtained daily from four women and four men (ages 24-49 years) over a period of 4-6 weeks. A lO-inL sample of the first morning void was collected, and the subjects kept a record of the amount of alcoholic beverages, if any, consumed the previous day throughout this period. The concentration of ethanol and the SHTOL/5HIAA ratio were determined. In an earlier study on the use of 5HTOL as a marker of recent drinking [24], the possible interference by metabolic acidosis in a subject with diabetes mellitus could not be ruled out. To examine this further, we assayed urine samples collected from 20 diabetic patients (12 women and 8 men, ages 38-91 years) admitted to St. G#{246}rans Hospital for treatment of ketoacidosis or coma. Thereafter, we assayed additional samples collected at 1-week intervals for a total of 3-5 weeks, having asked the patients not to consume any alcohol during this period. The concentrations of ethanol and acetone and the 5HTOL/5HIAA ratio were determined. All blood samples were collected from an antecubital vein into Vacutainer Tubes containing EDTA (Becton Dickinson, Meylan Cedex, France) for preparation of plasma by centrifugation. The urine samples were collected into plastic tubes without preservative. Before analysis, the plasma and urine samples were stored at -80 #{176}C. Samples were thawed overnight at 4 #{176}C. The study was approved by the local Ethical Committee (VSO/SSO) in Stockholm.

Results Figure 1 shows the time course of concentrations of ethanol in breath, plasma, and urine, and of methanol in plasma and urine, as well as the 5HTOL/5HIAA ratio in urine, for each of the 10 alcohol-dependent subjects during detoxification. Ethanol was usually detectable longer in urine than in breath or plasma. The concentration of methanol was markedly increased in the first sample [mean 0.57 ± 0.15 (SD) mmol/L in plasma and 0.62 ± 0.17 mmol/L in urine] and did not start to fall until the blood ethanol had returned to background concentration. Thereafter, methanol remained increased for a further 2-6 h (mean 3.5 h). The urinary SHTOL/51-HAA ratio in the first sample ranged between 331 and 1081 pmol/nmol (mean 684 ± 259 pmol/ nmol), and remained above the cutoff for 6-15 h (mean 10 h) after ethanol was no longer detectable in blood samples. Figure 2 shows the time course of concentrations of ethanol

620

Helander

Breath

0

et al.: Lab testing

D Urinary

Ethanol

Plasma Ethanol

80

Plasma

A

for alcohol

consumption

Ethanol

U Urinary

Methanol

Methanol

*

5HTOL/5HIAA

1.0

80

1.0

60

0.8

Urinary

#{149} #{182}000 E 100

:;

60

0.8

E E06

o

E 40 #{149}6 C #{149}

30

20

0.2

0

0

80

1.0

60

0.8

20 10

0 12

80

16

20

=

U,

-

24

1.0

#{149} #{182}000 6 #{149}I00

60

0.8

E E0.6

o

E 40

-

= #{149}

20 Ui

0

a so

0.6 40

0.4

20

.T1TTT7

. .,.

-

80

1000

20

0

0

80

1.0

80

60

30

02

0

#{182}000 -100 -

0

a 0.

40

40

40

20

20

=

0.4

30

C

Ui

0.6

50

U)

20

-j

02

0 I-

l0 0

0

80

60

1.0

E E0’6

E

100

40

0.4

A

80

1.0

60

0.8

20

0.2

U)

-

0

4

8

12

16

20

24

40

40 0.4

20

0.2

0

0

80

1.0

60

0.8

30

80

#{149}I0

0

E

a

0

;:I

40 0

40 0.4

C

= 20

20

0

0

30

02

20

UJ

to 0

4

8

12

#{182}620

24

ii,

#{149} iooo

0.6

40 0

U,

20

-J

a

a

28

80

0

a 0

to 0

0

1000 tOO

50

0.6

30

C

.c

0

0

50

-

= #{149} 20 uj

1000

-

0.8

o

.4Q

-

u,

0.8

0

0

3

10

100

E

-

40 0.4

_ .= o.z

0

-J ..

-

40 0.4

0A #{149}

0

=

E 50

40 -

20 Ui

0

0.6

0

0

28

Tim. (hI

_ = U, 0

v,

Tim. (hI

Ag. 1. Time course of concentrations of ethanol in breath, plasma, and urine, and of methanol in plasma and urine, as well as the ratio of 5HTOL/5HIAA in urine, in samples collected from 10 male alcohol-dependent subjects (A-fl during detoxification. The dashed line represents the cutoff limit used to indicate an abnormal 5HTOL/51-IIM ratio (>15 pmol/nmol). in plasma and urine, and of methanol in urine, as well as the 5HTOL/5HIAA ratio in urine, for the five female and five male social drinkers after administration of ethanol at 0.8 g/kg body weight. The concentration of methanol was highest in the 6-h sample (-5 times baseline) and decreased again to the baseline concentration (100-fold in the 6-h sample and did not return to baseline until after 14-16 h. No significant gender differences in ethanol elimination rate or in the urinary excretion patterns of methanol and 5HTOL/5HIAA were seen. In the experiment involving analysis of consecutive morning

Clinical Chemistry

P-ethanol

0

U-methanol

0 U-ethanol

U P-ethanol

9

0.20-

#{149} U-ethanol

9

A

U-methanol 9

42, No. 4, 1996

U-5HTOL/5HIAA #{149} U-5HTOL/5HIAA

621

g

1000

25

#{149}000E -J

0.15

‘..-.

-

0

20

-J 0

E

15

E 0.10

-

Ag. 2. Concentration-time

0 C

iO

(P) and urine (U), methanol

10

30 20

Ui

0.05

profiles of ethanol in plasma in urine, and the ratio of 5HTOL/5HIAA in Urine, in samples collected from healthy social-drinking volunteers (five women and five men, ages 23-39 years) after oral administration of ethanol at 0.8 g/kg body weight. The drink was finished within 30 mm in the morning after an overnight fast. Data represent mean values ± SD for the 5HTOL/

-

5

I-

10 0

0 0

2

6

10

14

18

5HIAA ratio, but, for clarity, SOs have been omitted for the other variables. The dashed line represents the cutoff limit used to indicate an abnormal 5HTOL/5HIAA ratio (>15 pmol/nmol(.

22

Time (h) urine samples, an average of 29 g of alcohol (range 3-98 g/day) had been consumed the previous day on 155 of a total of 261 sampling days, according to the self-reported drinking protocols. Usually, drinking occurred in the evening or late afternoon. Analysis of the urinary 5HTOL/5HIAA ratio in the first morning void identified on the average 73% (sensitivity) of all drinking occasions exceeding 7 g of alcohol, as compared with 22% by analysis of ethanol (>200 mol/L) (Table 1). On some occasions when relatively large amounts of alcohol had been consumed, the second and, in one case, even the third void of urine the next day confirmed that the 5HTOL/5HIAA ratio remained increased for considerably longer than ethanol could be detected (Fig. 3). The mean sensitivity for analysis of ethanol and the 5HTOL marker was somewhat less for women than for men (Table 1). However, two women reported that they occasionally voided during the night. The specificity was 100% for

Table 1. Sensitivity

and specIficity

both ethanol and 5HTOL/5HIAA in all subjects except one. This male subject showed one slightly increased metabolite ratio (21 pmol/nmol), although no alcohol consumption was reported the previous evening. We observed a significant correlation between the amount of alcohol consumed and the urinary 5HTOL/5HIAA ratio in the first morning void the next day (Fig. 4). In the urine samples collected from 20 diabetic patients with metabolic ketoacidosis or admitted to hospital in diabetic coma, the highest concentration of acetone was 4.4 mmol/L. The 5HTOL/5HIAA ratios were 6.5 ± 3.4 pmol/nmol (mean ± SD, n = 85), and all samples except one showed ratios below the cutoff limit of 15 pmol/nmol. This single value was only slightly increased (17 pmol/nmol) and was present in the urine sample collected immediately upon admission to the hospital.

of urinary (U) ethanol and 5HTOL/5HIAA

ratio to detect

recent alcohol consumption

in

healthy social drinkers. Increased No. of

days

Age,

Total

With >7 g

years

study

ethanol

43

23

38

21

27 35

21

5lITOL/5HIAA

U-ethanol >200

Alcohol intake range, g (mean)

No. days

Sens., %

9-98 (29.8) 8-48 (24.5) 8-74 (27.8) 14-83 (45.0)

20 14

20 29

Normal

amoi/L

U-ethanol gemol/L

5HIOL/5HIAAb No. of alcoholfree days

No. days

22

20

20

33

12

12

6

6

5

4

No. days

Sen..,

87

5

67

7

95

4

19

97

16

53

%

15 pmol/nmol. ‘15 pmol/nmol).

Discussion Hitherto, the determination of ethanol in body fluids or breath has represented the only generally available objective method to prove recent consumption of alcohol. However, because ethanol is rapidly eliminated from the body, this approach will only detect very recent drinking, even though ethanol can be detected for some hours longer in urine than in blood or breath, owing to retention of urine in the bladder. The use of even more sensitive analytical assays may increase diagnostic sensitivity, but also the risk of false-positive results, because several microorganisms occasionally found in humans are capable of producing ethanol by metabolic conversion of glucose or other endogenous substrates /25]. An increased concentration of methanol in body fluids or breath has been suggested as an alternative indicator of recent drinking [13]. The concentration of methanol increases during a drinking spree owing to competitive inhibition of alcohol dehydrogenase (ADH). The methanol might originate from endogenous sources, but it is also present as a congener in alcoholic beverages and can be derived from various foodstuffs (e.g., fruits and fruit juices) [26]. In the present experiments, methanol remained increased even after the blood ethanol concentration had returned to background concentrations in both alcoholdependent subjects and healthy volunteers. This time lag between ethanol and methanol clearance agrees with previous reports [14-17] and implies that methanol is more sensitive than ethanol as indicator of recent drinking. However, an abnormally high concentration of methanol in blood or urine could reflect recent as well as long-term continuous drinking. After acute intake of alcohol by healthy volunteers, the concentration of methanol increased to -0.14 mmol/L on average before returning to baseline values after 10-12 h. During chronic drinking, methanol accumulates gradually to reach much higher concentrations, and complete clearance after termination of drinking might require several days [14]. This was further confirmed in the alcohol-dependent subjects during detoxification, where methanol leveled out within the concentration range reached in

0

20

40

60

80

100

Amount of alcohol (g) Fig. 4. Correlation between the urinary 5HTOL/5HIAA ratio in the first morning void and the amount of alcohol consumed (range 3-98 g) on the day before sampling in eight social-drinking female and male subjects. Accordingto the self-report protocols, drinking usually occurred in the evening or late afternoon. The dashed line represents the cutoff limit used to indicate an abnormal 5HTOL/5HIAA ratio (>15 pmol/nmol).

social drinkers after the ingestion of ethanol at 0.8 g/kg body weight. Accordingly, an increased methanol concentration is not specific for recent alcohol consumption, but may be useful for detection of chronic drinking [27, 28]. The results of the present study demonstrate that the urinary 5HTOL/5HJAA ratio provides a more sensitive method to detect recent alcohol consumption compared with analysis of ethanol or methanol in blood, breath, or urine. The specificity for detecting alcohol was also very high with the 5HTOL/ 5HIAA ratio [29]. A dose-dependent increase in the urinary excretion of 5HTOL is seen after each drink, owing to competitive inhibition of ALDH by ethanol-derived acetaldehyde and (or) the change in hepatic redox state [30]. In contrast to methanol, the baseline value for 5HTOL/5HIAA is not increased after prolonged misuse and can therefore be used to identify recent drinking in both social drinkers and chronic consumers. The major advantage of 5HTOL/5HIAA compared with ethanol and methanol is that the serotonin metabolite ratio stays increased for several hours longer than ethanol and methanol, thereby increasing diagnostic sensitivity. However, the exact time available for detection of acute alcohol consumption by analysis of urine depends not only on the dose and time since the last drink, but also on voiding between the time of drinking and sampling. To enhance the specificity of the marker, 5HTOL is expressed as a ratio with 5HIAA rather than creatinine, because dietary serotonin might otherwise cause false-positive results owing to a general increase in the urinary output of serotonin metabolites [20]. To discriminate between normal and increased values of the urinary 5HTOL/5HIAA ratio, the cutoff limit was originally set at 20 pmollnmol [12]. However, after careful study of a large number of abstinent subjects of both Caucasian and Oriental origin [22/, a lower limit of 15 pmollnmol was introduced into clinical practice. Actually, all those subjects had ratios

Clinical Chemistry

urinary or plasma methanol > urinary ethanol > blood or breath ethanol. Measuring 5HTOL/5HIAA can detect consumption of even moderate amounts of alcohol within the preceding -24 h and, by frequent urine sampling, a detailed picture of a person’s drinking pattern can be obtained. This new biochemical marker can be useful for investigating single relapses in alcoholic outpatients, to prove a person actually has abstained from alcohol [35, 36], and possibly in connection with workplace alcohol testing, e.g., as a postaccident test of recent heavy drinking (possible impairment caused by hangover). Furthermore, 5HTOL is a much more objective measure than self-reported alcohol consumption when new treatment models and new biochemical markers of alcohol misuse are being evaluated. Further research should focus on the clinical usefulness of 5HTOL/5HIAA measurements for monitoring relapse in drinking.

1987;24:321-32.

16. Jones AW, Sternebring B. Kinetics of ethanol and methanol in alcoholics during detoxification. Alcohol Alcohol 1992:27:641-7. 17. Haffner H-T, Batra A, Wehner HD, Besserer K, Mann K. Methanolspiegel und Methanolelimination bei Alkoholikern. Blutalkohol 1993:30:52-62.

Jones AW, Lovinger H. Relationship between the concentrations of ethanol and methanol in blood samples from Swedish drinking drivers. Forensic Sd Int 1988:37:277-85. 19. Jones AW, Schuberth J. Computer-aided headspace gas chromatography applied to blood-alcohol analysis: importance of on-line process control. J Forensic Sci 1989:34:1116-27. 20. Helander A, Wikstr#{228}m T, L#{228}wenmo C, Jacobsson G, Beck 0. Urinary excretion of 5-hydroxyindole-3-acetic acid and 5-hydroxytryptophol after oral loading with serotonin. Life Sci 1992: 18.

50:1207-13.

This work was supported in part by the Karolinska Institute. We thank Ulla Lindstr#{246}mfor the collection of samples, and Tina Bur#{233}nius, Gun Jacobsson, and Catharina L#{246}wenmo for skillful technical assistance.

References Fuller RK, Lee KK, Gordis E. Validity of self-report in alcoholism research: results of a Veterans Administration Cooperative Study. Alcohol Clin Exp Res 1988;12:201-5. 2. Ness DE, Ende J. Denial in the medical interview: recognition and management. JAMA 1994:272:1777-81. 3. Nilssen 0, Huseby NE, H#{248}yer G, Brenn T, Schirmer H, F#{248}rde OH. New alcohol markers-how useful are they in population studies: the Svalbard Study 1988-1989. Alcohol Clin Exp Res 1992:16: 1.

82-6.

4. Mihas AA, Tavassoli M. Laboratory markers of ethanol intake and abuse: a critical appraisal. Am J Med Sci 1992;303:415-28.

Helander A, Beck 0, Wennberg M, Wikstr#{246}m T, Jacobsson G. Determination of urinary 5-hydroxyindole-3-acetic acid by highperformance liquid chromatography with electrochemical detection and direct sample injection. Anal Biochem 1991:196:170-3. 22. Helander A, Walzer C, Beck 0, Balant L, Borg S, von Warthurg J-P. Genetic variation in alcohol dehydrogenase and aldehyde dehydrogenase and serotonin metabolism. Life Sci 1994;55:359-66. 23. Helander A, Beck 0, Borg S. The use of 5-hydroxytryptophol as an alcohol intake marker. Alcohol Alcohol 1994:2(Suppl):497-502. 24. Helander A, Beck 0, Jones AW. Distinguishing ingested ethanol from microbial formation by analysis of urinary 5-hydroxytryptophol and 5-hydroxyindoleacetic acid. J Forensic Sd 1995:40:95-8. 25. Saady JJ, Poklis A, Dalton HP. Production of urinary ethanol after sample collection. J Forensic Sci 1993:38:1467-71. 26. GrUner 0, Bilzer N. Zum Methanolgehalt von Fruchts#{228}ften. Seine Bedeutung bei der Begleitstoffanalyse. Blutalkohol 1983:20: 241-52. 27. Rome RP, Eriksson CJP, Ylikahri R, Penttil#{228} A, Salaspuro M. 21.

624

Helander

Methanol

as a marker of alcohol abuse.

et al.: Lab testing

Alcohol Clin Exp Res

1989:13:172-5.

28.

Iffland R, Balling P, Borsch G, Herold C, Kaschade

W, LOffler T, et Aceton Kraftfahrer. Blutalkohol

al. Zur Wertung erh#{246}hter Spiegel von GGT, CDT, Methanol,

29.

30. 31.

32.

und Isopropanol im Blut alkoholauffalliger 1994;31:273-314. Helander A, Beck 0, Boysen L. 5-Hydroxytryptophol conjugation in man: influence of alcohol consumption and altered serotonin turnover. Life Sci 1995:56:1529-34. Walsh Mi. Role of acetaldehyde in the interactions of ethanol with neuroamines. Adv Mental Sci 1973:3:233-66. Helander A, Beck 0, Borg S. Determination of urinary 5-hydroxytryptophol by high-performance liquid chromatography with electrochemical detection. J Chromatogr 1992:579:340-5. Thomasson HR, Crabb DW, Edenberg Hi, Li T-K. Alcohol and aldehyde dehydrogenase polymorphism and alcoholism. Behav Genet 1993:23:131-6.

for alcohol

consumption

Beck 0, Helander A, Carlsson S. Borg S. Changes in serotonin metabolism during treatment with the aldehyde dehydrogenase inhibitors disulfiram and cyanamide. Pharmacol Toxicol 1995;77: 323-6. 34. Levy MS, Livingstone BL, Collins DM. A clinical comparison of disulfiram and calcium carbamide. Am J Psychiatry 1967;123: 1018-22. 35. Voltaire-Carlsson A, Hiltunen AJ, Beck 0, Stibler H, Borg S. Detection of relapses in alcohol-dependent patients: comparison 33.

of carbohydrate-deficient

in urine, 703-8. 36.

and

transferrin

self-reports.

in serum,

5-hydroxytryptophol

Alcohol Clin Exp Res

1993:17:

Helander A, Voltaire-Carlsson A, Borg S. Longitudinal comparison of carbohydrate-deficient transferrin and gamma-glutamyl transferase: complementary biochemical markers of excessive alcohol consumption.

Alcohol

Alcohol

1996;31:lOi.-7.