Urinary Excretion of Methanol and 5-Hydroxytryptophol

as Biochemical Markers of Hangover

A.W. Jones,1 P. Bendtsen,2 and A. Helander3

‘Department o f Forensic Toxicology and 2Drug Dependence Unit, University Flospital,

Linkoping, and ’Department o f Clinical Neuroscience, St Gorans Hospital, Stockholm,

Sweden.

INTRODUCTION

Besides the acute effects of ethanol on a person's performance and behavior, the after-effects

of heavy drinking, commonly referred to as the hangover state, causes impairment o f mental

and body functioning, such as fatigue, attention deficit, and diminished psychom otor skills

(Kelly et al., 1970; Seppala et al., 1976; Lemon et al. 1993). It seems likely that many

accidents on the roads and in the workplace might be attributed to decreased attention and

slower reaction time associated with the residual effects o f an evening’s heavy drinking

(Karvinen et al., 1962; Tornros and Laurell 1991; Roehrs et al., 1994). Although neither

dose-response relationships nor the prevalence of hangover have been clearly established,

experience has shown that a person’s blood alcohol concentration (BAC) should exceed

approximately 1.0 g/L during the acute phase of intoxication (Smith et al., 1983; Collins and

Chiles 1980). The hangover starts to develop when the BAC decreases again and is

approaching zero. However, the intensity and duration of hangover shows large inter- and

intra-individual variations and the reduced performance in some people might persist for

many hours after all ethanol has been cleared from the body (Goldberg, 1961).

BIOCHEM ICAL MARKERS OF HANGOVER

Low concentrations o f ethanol and methanol (0.5-1.0 mg/L) can be determined in body fluids

even without drinking alcoholic beverages (Majchrowicz and Mendelson, 1971). These

endogenous alcohols are oxidized in the liver to their respective aldehydes by alcohol

dehydrogenase (ADH) and the affinity o f this enzyme is roughly 10 times higher for ethanol

-255-

as its substrate compared with methanol. This means that the concentrations o f m ethanol in

blood and other body fluids increase during the time people drink alcoholic beverages

because the enzyme ADH is fully engaged in the metabolism of ethanol (M ajchrowitz and

Mendelson, 1971). Methanol occurs as a congener in alcoholic beverages, which also

contributes to the raised concentrations observed in body fluids after a drinking spree

(Majchrowitz and Mendelson, 1971). The elimination half-life o f methanol in humans is

between 2-4 hours (Jones 1987; Haffner et al. 1992) so that methanol can be detected in

blood, breath, and urine for long after ethanol has returned to its endogenous concentration

(Jones, 1986).

Acetaldehyde is the primary metabolite o f ethanol oxidation and this noxious substance is

rapidly converted to acetate through the action of aldehyde dehydrogenase (ALDH).

However, ALDH is also involved in the oxidation of biogenic aldehydes such as those formed

during catabolism of dopamine, noradrenaline and serotonin (5-hydroxytryptamine). W ithout

drinking any alcohol, the urinary excretion products o f serotonin m etabolism are 5-

hydroxyindoleacetic acid (5HIAA) and 5-hydroxytryptophol (5HTOL), and the ratio o f

5HTOL/5HIAA is normally very low (Helander et al., 1993). W hen acetaldehyde derived

from ethanol competes with the biogenic aldehyde formed from serotonin for available

ALDH, there is a switch in the metabolic pathway and the ratio 5HTOL/5HIAA in the urine

increases appreciably and does not recover to baseline levels until after ethanol has been

cleared from the blood (Helander et al., 1993).

Analyzing methanol and the metabolites o f serotonin (5HTOL and 5HIAA) in urine therefore

furnishes a way to disclose recent drinking even after ethanol has been cleared from the body

(Helander et al., 1996; Jones and Helander 1996).

CLINICAL AND FORENSIC APPLICATIONS

During rehabilitation of alcoholics and others who are forbidden to drink alcohol as part of

their treatment, a sensitive and specific test o f drinking is often necessary. Breath-ethanol

testing alone is not sufficiently sensitive to control whether a patient has remained abstinent

because many people can regulate their intake of alcohol. Studies have shown that the

analysis o f methanol and 5HTOL/5HIAA can serve as markers o f relapse and as a way to

-256-

m onitor whether a person has recently consumed alcohol. This is useful in connection with

the treatment o f recidivist drunk drivers who are not allowed to drink as a condition of

relicensing. During medicolegal investigations into the likely cause o f serious accidents in the

workplace it is obviously important to establish whether or not recent consumption of alcohol

was a factor involved. This might be relevant in civil and criminal litigation when

responsibility for the accident and insurance claims are made (Ames et al., 1997).

We have conducted a controlled drinking experiment to investigate the usefulness of

methanol and 5HTOL/5HIAA as biochemical markers of alcohol ingestion and to test how

long after the end of drinking these markers remain valid compared with measuring ethanol in

blood, breath, and urine.

EXPERIM ENTAL

Twenty healthy subjects (9 women and 11 men) with a mean age of 39 ± 9.4 y (± SD) and

mean body weight 72 ± 11.6 kg (± SD) participated in this study as paid volunteers. They

drank either 80 g or 50 g o f ethanol according to choice as white table wine (11% v/v) and

export beer (5.2 % v/v). The alcohol was consumed over a period of 2 hours together with a

meal in an attempt to mimic social drinking conditions. Fifteen minutes after the end of

drinking, breath-alcohol tests were made with an Alcolmeter S-D2 and the results were

reported as equivalent blood-alcohol concentration. Thereafter, a specimen of urine (10 ml)

was collected into a plastic tube containing 100 mg sodium fluoride as preservative. Another

breath-alcohol test was made and a specimen of urine collected just before bedtime. Breath-

alcohol concentration was measured in the morning after arising from bed and a sample of

urine was also collected. Breath and urine samples were taken on two further occasions

during the m orning or early afternoon. The times for collecting the consecutive samples of

urine were not exactly the same for each subject. Furthermore, a control specimen of urine

was either collected before ingestion o f alcohol began or on a separate occasion when the

subjects had abstained from drinking alcohol.

The concentrations o f ethanol and methanol in urine were determined by headspace gas

chromatography as reported in more detail elsewhere (Jones and Schuberth, 1989; Jones and

Lowinger 1988). For the analysis o f ethanol, the urine was diluted 1 + 10 with n-propanol

-257 -

(0.8 g/L) as an internal standard and the peak height ratios ethanol/n-propanol were used for

quantitative analysis by comparison with the response from analysis of known strength

alcohol standards (0.50, 1.00 and 1.50 g/L). The limit o f quantitation of the method is 0.01

g/L and the within-run precision expressed as coefficient of variation is 1% at a mean urine-

ethanol concentration of 1.0 g/L. For the analysis of methanol a salting-out procedure was

used to enhance the analytical sensitivity. This entailed adding the aliquot of urine (0.5 ml) to

a headspace vial containing 1.2 g sodium chloride. The vials (22 ml volume) were

immediately made air-tight with crimped-on rubber septums before being analyzed. The

precision (CV) of analyzing methanol in urine by this method was 3.4 % at a mean

concentration of 2.3 mg/L.

The concentrations o f 5-hydroxytryptophol (5HTOL) and 5-hydroxyindoleacetic acid

(5HIAA) in urine were determined by gas chromatography-mass spectrometry (GC-MS) and

high performance liquid chromatography (HPLC), respectively. These methods have been

validated and described in detail elsewhere (Beck et al., 1982; Helander et al., 1991). To

compensate for variations in urine flow rate and also dietary intake o f serotonin, which are

factors that might influence results, the ratio o f 5HTOL/5HIAA was calculated.

The concentrations of methanol and the ratio 5HTOL/5HIAA in the pre-drinking specimens

of urine were compared with results for samples collected after drinking by a Student’s t-test

for paired observations (dependent t-test).

RESULTS

Figures 1-3 show the concentrations of ethanol, methanol, and the ratio 5HTOL/5HIAA in

urine samples collected before drinking alcohol and on five occasions after intake of either 50

g or 80 g ethanol. As expected, the UAC, UMC, and the ratio o f 5HTOL/5HIAA were higher

after the subjects had consumed 80 g ethanol compared with 50 g.

In the morning after drinking 50 or 80 g ethanol, the breath-test results were zero for all

except three individuals who registered 0.01, 0.05, and 0.09 g/L on the Alcolmeter S-D2

device. The highest BrAC was obtained for the person with highest morning UAC of 0.76

g/L. The first morning voids as a rule contained measurable amounts o f ethanol, being 0.38 ±

-258 -

0.10 g/L after drinking 80 g ethanol and 0.09 ±0.03 g/L after intake of 50 g ethanol.

However, ethanol was not measurable in urine (< 0.01 g/L) at the time of the second and third

voids obtained later in the morning or early afternoon, indicating alcohol-free status in all

subjects by this time.

Figures 2 and 3 show that urinary methanol and the ratio o f 5HTOL/5HIAA follow a

different time course compared with UAC, being shifted in time. This time lag is particularly

evident for urinary methanol, which peaks in the first morning void when the BrAC had

already reached zero in most subjects. Table 1 compares the statistical significance of

differences between the pre-drinking levels o f methanol and the 5HTOL/5HIAA ratio with

values determined in urine specimens collected during the morning and afternoon ( 1 st, 2 nd

and 3rd voids). The results for the second and third morning voids have practical importance

as a test for recent drinking because by the time of sampling ethanol was no longer detectable

in blood, breath, or urine and the hangover was most troublesome.

Figure 1. Concentrations of ethanol in urine after subjects drank 50 g (N = 11) or 80 g

(N = 9) ethanol as beer and white wine.

-259 -

Figure 2. Concentrations of methanol in urine collected before and after intake o f 50 g

(N = 11) or 80 g (N = 9) ethanol as beer and white wine.

Figure 3 : Concentration ratios o f 5HTOL/5HIAA (pmol/nmol) in urine samples

collected before and after intake of alcohol 80 g (N = 9) or 50 g (N = 11) in the form of

white wine and beer.

-260-

Table 1: Differences between urinary concentrations o f methanol (mg/L) and the ratio

of 5H TOL/5HIAA in pre-drinking specimens of urine and voids made in the morning

after an evening’s heavy drinking. Statistical testing was by Student’s t-test for paired

observations, where; * p < 0.05, ** p<0.01, *** p<0.001.

Analyte EtOHDose

Urine-4 1 st morning void

Urine-5 2 nd morning void

Urine - 6

3rd morning void

Methanol 50 g 3.06 ± 0.40**5.06 ±0.55***

0.73 ± 0.16** 2.28 ±0.67**

0.26 ± 0.09* 0.49 ± 0 .3 2

5HTOL/5HIAA Uloo

O 134 ± 28.7*** 296 ±47.3***

4.6 ± 1.23** 35.0 ±8.41**

1.54 ± 0.41* 6 . 2 2 ± 2.26*

The first m orning void was collected between 7-10 hours after bedtime, the second morning

void 3-6 hours later and the third and last void between 5-8 hours later. However, although

concentrations of methanol and the 5HTOL/5HTOL ratio could be distinguished from the

pre-drinking values, the last specimen of urine (3rd morning void) would not be o f much

practical use without knowledge of a person’s pre-drinking levels. This information is mostly

not available.

DISCUSSION

It is common knowledge that people do not function normally when they are suffering from a

hangover and this can be demonstrated when skilled tasks must be performed (Kelly et al.,

1970; Seppala et al., 1976; Lemon et al. 1993). For this reason ingestion of alcohol is strictly

regulated for those engaged in safety-sensitive work (Dubowski and Kaplan, 1996).

Nevertheless, too m uch drinking leads to many accidents and premature deaths and minimum

periods o f alcohol deprivation have been introduced for those engaged in highly demanding

tasks such as aviation pilots. Until now the hangover effects o f too much drinking have not

been seriously considered as a cause o f accidents because reliable markers o f alcohol

consumption have not been available after ethanol has been metabolized. W hen investigating

the causes o f serious accidents alcohol and drug use should always be considered as likely

contributing factors. However, a low blood or urine alcohol concentration at autopsy, e.g. less

than 0.10 g/L, is not considered relevant and previous alcohol ingestion and intoxication are

-261 -

usually written off as likely causes o f the accident. However, an autopsy BAC of 0.01 g/L

might reflect a BAC of 2.0 g/L the previous evening and the person concerned might

therefore have been suffering from a hangover when the accident happened. Sensitive and

specific biochemical tests of recent drinking even after ethanol has been cleared from the

body would be useful in medicolegal investigations o f fatal accidents. In this study, most

of the subjects had eliminated all alcohol from the blood sometime during the night so that

the UAC in the first morning void reflects the BAC prevailing since the bladder was last

emptied, that is, before bedtime. The concentrations of methanol and the 5HTOL/5HIAA

ratio were elevated above pre-drinking baseline values in the first, the second, and sometimes

even the third urine voids collected during the morning and early afternoon and therefore up

to 5-10 hours after BrAC was no longer measurable.

Some investigators advocate measuring methanol in body fluids as a marker for chronic

consumption of alcohol and alcoholism (Iffland, 1996). If the concentration is above 10

mg/L in serum this is considered a marker of heavy and continuous drinking for days or

weeks. By contrast, we are advocating the analysis of methanol in urine as a marker o f acute

intake of alcohol and this test is sufficiently sensitive to disclose ingestion of alcohol even

after alcohol has been cleared from the body. However, the analysis o f methanol alone is not

completely specific as a test for alcohol consumption because eating fresh fruits and drinking

large volumes of fruit juice containing traces o f free methanol or methyl esters can account

for the urinary excretion of this alcohol. Even the artificial sweetner aspartame can be

converted in the body to methanol (Davoll et al., 1986). Accordingly an elevated

concentration of methanol in urine needs to be supported by other evidence or a more specific

marker o f alcohol consumption such as an elevated ratio o f 5HTOL/5HIAA.

These new biochemical markers (methanol and 5HTOL) furnish a way to disclose recent

drinking for up to 5-10 hours after alcohol is no longer measurable by conventional breath-

alcohol tests. There will certainly be applications as "relapse m arkers” during rehabilitation of

alcoholics and drug addicts as well as convicted drunk drivers who must refrain from

drinking as a part of their treatment. Finding an elevated urinary concentration of methanol

and an elevated ratio o f 5HTOL/5HIAA provides convincing evidence of recent alcohol

consumption even if ethanol cannot be determined in body fluids. Because a person’s pre­

drinking levels of urinary methanol and the ratio 5HTOL/5HIAA are not usually known, we

-262-

consider that threshold values of 2.0 mg/L for methanol and 15 pmol/nmol for

5HTOL/5HIAA are reasonably values for clinical and forensic purposes when evaluating

random samples o f urine.

REFERENCES

Ames, G.M., Grube, J.W., and Moore, R.S. (1997). The relationship of drinking history and hangovers to workplace problems: An empirical study. J. Stud. Ale. 58, 37-47.

Beck, O., Borg, S., Eriksson, L., Lundman, A., (1982). 5-Hydroxytryptophol in the cerebrospinal fluid and urine o f alcoholics and healthy subjects. Naunyn-Schmeidebergs Arch Pharmacol 321;293-297.

Collins, W .E., and Chiles, W.D. (1980). Laboratory performance during acute alcoholintoxication and hangover. Human Factors 22, 445-462.

Davoll, L., Cappellini, L., Airold, L., and Fanelli, R., (1986). Serum methanol concentrations in rats and in men after a single dose o f aspartame. Fd Chem Tox 24; 187-189.

Dubowski, K.M., and Kaplan, Y. (1996). Alcohol testing in the workplace. Chapter 19 in Medicolegal aspects o f alcohol, edited by J.C. Garriott, Lawyers & Judges Publishing Co., Tucson, pp 439-475.

Goldberg, L. (1961). Alcohol, tranquilizers and hangover. Quart J Stud A le Suppl 1 37-56.

Haffner H.T., W ehner, H.D, Scheytt, K.D., and Besserer, K., (1992). The elimination kinetics of methanol and the influence of ethanol. In t J Leg M ed 195; 111-114.

Helander, A., Beck, O., W ennberg, M., W ikstrom, T., and Jacobsson, G. (1991).Determination of urinary 5-hydroxyindole-3-acetic acid by high performance liquid chromatography with electrochemical detection and direct sampling injection. Anal Biochem 196:170-173.

Helander, A., Beck, O., Jacobsson, G., Lowenmo, C., and Wikstrom, T. (1993). Time course o f ethanol-induced changes in serotonin metabolism. Life Sci 53;847-855.

Helander, A., Beck, O., and Jones, A.W. (1996). Laboratory testing for recent alcohol consumption; comparison of ethanol, methanol, and 5-hydroxytryptophol. Clin Chem 42;618-624.

Iffland, R. (1996). New ways to use biochemical indicators o f alcohol abuse to regrant licences in a fairer manner after drunken driving in Germany. Ale. Alcohol. 31, 619-620.

Jones, A.W. (1986). Abnormally high concentration of methanol in breath; A useful biochemical marker o f recent heavy drinking. Clin Chem 32; 1241-1242.

Jones, A.W. (1987). Elimination half-life o f methanol durimg hangover. Pharmacol. Toxicol.

-263 -

60. 217-220.

Jones, A.W., and Lowinger H. (1988). Relationship between the concentrations of ethanol and methanol in blood samples from Swedish drinking drivers. Forens Sci Intern 37;277-285.

Jones, A.W., and Schuberth, J. (1989). Computer-aided headspace gas chromatography applied to blood-alcohol analysis; Importance of on-line process control. J Forens Sci 34; 1116-1127.

Jones, A.W. and Helander, A. (1996). Disclosing recent drinking after alcohol has been cleared from the body. J. Anal. Toxicol. 20, 141-142.

Karvinen, E., Miettinen, M., and Ahlman, K. (1962). Physical perform ance during hangover. Quart J Stud A le 23, 208-215.

Kelly, M., Myrsten, A.L., Neri, A., and Rydberg, U. (1970). Effects and after-effects of alcohol on physiological and psychological functions in man - a controlled study. Blutalkohol 7, 422-436.

Lemon, J., Chesher, G., Fox A., Greeley, J., and Nabke, C., (1993). Investigation o f the «hangover» effects o f an acute dose of alcohol on psychomotor performance. A lcohol Clin Exp Res 17, 665-668.

M ajchrowicz, E. and Mendelson, J.H. (1971). Blood methanol concentrations during experimentally induced ethanol intoxication in alcoholics. J. Pharmacol. Exp. Ther. 179, 293-300.

Roehrs, T., Claiborue, D., Knox, M. and Roth, T. (1994). Residual sedating effects of ethanol. Alcohol Clin. Exp. Res. 18, 831-834.

Seppala, T., Leion, T., Linnoila, M., Huttunen, M., and Yilikahri, R. (1976). Effects of hangover on psychomotor skills related to driving: Modifications by fructose and glucose. Acta Pharmacol Toxicol 38, 209-218.

Smith C.M., and Barnes, G.M. (1983). Signs and symptoms of hangover: Prevalence and relationship to alcohol use in a general adult population. Drug A le Abuse 11, 249-269.

Tornros, J., and Laurell, H. (1991). Acute and hangover effects o f alcohol on simulated driving performance. Blutalkohol 28, 24-30.

-264-