Schaffer Online Library of Drug Policy Sign the Resolution
Contents | Feedback | Search
DRCNet Home
| Join DRCNet
DRCNet Library | Schaffer Library | Drugs and Driving 


Experiences with Volatile Alcoholism Indicators (Methanol, Acetone, Isopropanol) in DWI Car Drivers

R. Iffland, G. Berghaus

Institute of Forensic Medicine, University of Cologne, Melatengürtel 60-62, 50823 Cologne, Germany


State markers of alcoholism can be divided into (1) the indicators connected directly to ethanol metabolism and the substances involved in or by ethanol metabolism, and (2) the indicators that are not associated with actual blood alcohol concentrations, such as GGT, CDT, or MCV. Methanol, acetone, and isopropanol belong to the first group. These substances were detected and proposed as alcoholism indicators in connection with congener alcohol analyses of blood samples from DWIs. Methanol, in particular, has been adopted as a powerful indicator of alcohol misuse. Levels greater than 10.0 mg/kg blood indicate, in most cases, addictive drinking of long duration.

Acetone and isopropanol are bound by a redox-reaction. During the storage of blood samples, acetone can be altered to isopropanol. Therefore, it is reasonable to adopt the combination of acetone + isopropanol as a new alcoholism indicator, with 9.0 mg/kg as the cut-off. Experiences with these volatile alcoholism indicators are reported in re-analysis of blood samples obtained from intoxicated drivers.


Blood samples are obligatory in Germany at drunken driving to estimate blood alcohol concentrations (BAC). Jurisdiction affords high accuracy in determination to prevent misjudgment. Congener analyses of blood samples are carried out in cases of claiming after drinks to prove these answers. The concentrations of these alcohols like methanol, n-propanol or isobutanol in blood compared with ethanol are very low. They don't exceed hardly 1 mg/kg except of methanol, which may reach levels of 20 mg/kg. Acetone and isopropanol are physiological substances with similar blood levels. They are measured by the same gaschromatographic method like the congener alcohols. These analyses are usual since the beginning of the eighties in many German institutes of forensic medicine but not in other parts of the world. Most of the papers about this new field of research are written in German. This and different systems in justice are the main reasons of the poor knowledge in English spoken countries about German alcohol investigations.


Nearly all alcoholic beverages contain methanol (Table 1). In contrast to n-propanol or isobutanol methanol of these beverages is completely resorbed on account of the total inhibition of methanol metabolism by ethanol. Methanol blood levels exceed however the expected concentrations. Also if dinking is stopped and no more ethanol resorbed, the blood levels of methanol increase until ethanol concentrations are fallen down to nearly 0.2 g/kg (Iffland et al. 1994). This is only to explain by endogenous sources of methanol. Majchrowicz and Mendelson (1971) were the first, who described this phenomenon.

Table 1
Methanol and Ethanol Contents of Alcoholic Beverages

Beverage Methanol (mg/l) Ethanol (g/l)
Beer 1 - 10 30 - 50
White Wine 20 - 40 60 - 100
Red Wine 60 - 100 70 - 110
Brandy 200 - 300 300
Vodka 1 - 100 300 - 400
Whisky 80 - 200 300 - 320
Fruiterer 1000 - 4000 300 - 350

Methanol levels of 0.5 - 1.0 mg/kg exist in humans already in sober condition. During drinking most methanol is formed endogenous. Levels of 0.2 - 0.3 mg/kg are formed hourly as long as blood ethanol concentrations exceed 0.2 g/kg. The source of this methanol is not found till now. There exist some hypotheses like reduction of formic aldehyde built by methyl donators, bacterial formation of methanol in intestine or splitting of methyl esters like pectines in food. But these are no explanations for a constant endogenous formation of methanol. Tiess and Stöhlmacher (1990) assume a minor metabolic way of ethanol forming small amounts of methanol. This hypothesis is supported by drinking trials with isopropanol which also inhibits methanol metabolism at the ADH. But no increasing methanol levels like at ethanol were observed in these attempts (Iffland et al. 1989).

First hints to methanol as an indicator of alcohol abuse were given by Iffland et al. (1984). They found more frequent methanol levels higher than 10 mg/kg in the blood of alcoholics, who were hospitalized drunken, or people known as alcoholics, who died alcoholized. The main source of blood methanol was in these cases endogenous methanol for it is impossible to reach such high methanol levels by the most alcoholic beverages except by fruiterers. High methanol levels require a permanent alcoholization of more than one day and continuing drinking on the next day not interrupted by sobriety. This behaviour characterizes an addicted drinking and makes methanol to a potent indicator of alcohol misuse.


Crato et al. (1978) described first enlarged acetone blood levels > 7.0 mg/kg in cases of blood ethanol concentrations higher than 2.0 g/kg. Acetone is formed by decarboxylation of aceto-acetic-acid which has acetyl-coenzyme A as precursor. Increased formation of NADH - directly connected to ethanol metabolism - delays degradation of acetyl coenzyme A and enlarges on this way acetone blood levels. Condensation of pyruvate and acetaldehyde is seen as an explanation of acetone formation too. Metabolism of ethanol is increased at blood levels > 2.0 g/kg and connected thereby with higher amounts of NADH and acetaldehyde the reasons to form higher amounts of acetone. Acetone blood levels > 7.0 mg/l are seen as indicators for alcohol misuse (Iffland et al. 1994). Alcoholic ketoacidosis is the final form of ethanol induced enlarged endogenous formation of ketones in humans (Caspar et al. 1993).


Isopropanol is metabolized by hepatic ADH to acetone like ethanol to acetaldehyde. The balance of these reactions is shifted in the liver on the side of the alcohols. In contrast to acetaldehyde there exists no further metabolism of acetone. This explains the observations of Tiess (1985) that in liver slices isopropanol was formed from acetone. Decarboxylation of b-hydroxybutyracid is supposed as another way similar to acetone forming isopropanol. This narrow connection to ketones makes also isopropanol to an important indicator of alcohol abuse. Normal physiological blood levels of isopropanol don't exceed 0.1 - 0.2 mg/kg. Isopropanol was proposed as an indicator of alcohol misuse first by Iffland et al. (1989, 1994). The cut-off was set to 2.0 mg/kg.

The narrow connection of acetone and isopropanol suggests to summarize both indicators and to define it as a new marker, AcIp = acetone+isopropanol. During storage of blood samples parts of acetone undergo a reduction to isopropanol. The new marker AcIp is for reanalyses of blood samples appropriate. The cut-off was set to 9.0 mg/kg (Krambrich 1993, Iffland et al. 1994).


Four different indicators of alcohol abuse - GGT, CDT, methanol and acetone+isopropanol - were determined in blood samples of 534 male alcoholized car drivers representative for German conditions (Iffland and Grassnack, 1995). The men were at least 18 years old and had minimum blood ethanol levels of 0.80 g/kg. Included blood alcohol concentration five indicators were measured in every case characteristic for different drinking behaviour (Table 2)(Iffland et al. 1994).

Table 2
Discrimination of Drinking Behaviour by High Levels of Indicators of Alcohol Abuse

BAC Acute drinking of large amounts of alcohol
Methanol Consuming alcohol without time of sobriety
AcIp Alcohol induced disturbances of metabolism
CDT Drinking alcohol (> 100 g/d) for some weeks
GGT Drinking large amounts of alcohol for years

The distributions of methanol and AcIp are similar in this study and don't exceed most of all 5.00 mg/kg. Cases with higher levels were found at methanol 121 (22.7 percent) and at AcIp 64 (12.0 percent). Blood concentrations of methanol are normal up to 5.00 mg/kg and may be reached after consuming 60 - 100 g ethanol with alcoholic beverages in some hours for example during an evening with good friends. Levels between 7.00 and 9.99 mg/kg are suspicious for alcohol abuse. They were found in 37 (6.9 percent) car drivers and in 30 (5.6 percent) alcohol abuse is sure. AcIp exceeded in 17 cases (3.2 percent) the level of 9.00 mg/kg and laid at 27 (5.1 percent) between 6.00 and 8.99 mg/kg comparable with the group of suspicious alcohol abuse.

There exist no functional but sometimes high stochastic correlations between methanol (MeOH), AcIp and other parameters in this study. Ranking correlations were calculated for two groups. Group A contains all cases (534) and group B car drivers (104) with BAC from 2.00 g/kg (Table 3). The highest correlations are between MeOH and AcIp. The coefficients of methanol to GGT and BAC are remarkably high in group B. Poor correlation exists to CDT. This indicator seems to be most independent from other indicators.

Table 3
Ranking Correlation Coefficients of the Indicators

Methanol A 0.1308 0.1706 0.1568 0.5119 1.0000
B 0.4018 0.4143 0.1646 0.6394 1.0000
AcIp A 0.1998 0.3945 0.2365 1.0000 0.5119
B 0.2925 0.4900 0.2774 1.0000 0.6394

It is important to know, who of the drivers has heavy problems with alcohol. Important hints give the levels of these indicators. It was differed between three stages (Table 4). The cut-offs of heavy alcohol misuse were set for GGT at 70 U/1 (25°C) and for CDT at 60.0 U/1 (RIA-method, Kabi Pharmacia Diagnostics).

Table 4
Levels of the Indicators as Criteria of the Extent of Alcohol Abuse

Alcohol abuse BAC
No proof < 1.60 < 7.00 < 6.00 < 30.0 < 40
Possible 1.60 - 2.49 7.00 - 9.99 6.00 - 8.99 30.0 - 59.9 40 - 69
Heavy > 2.49 > 9.99 > 8.99 > 60.0 > 70

The cut-offs marking heavy problems were exceeded by at least one of the five parameters in 116 cases (21,7 percent). Two or more parameters exceeded these limits in 37 cases. But the criterion for alcohol abuse is satisfied, if only one of the five parameters is higher than the cut-off. 8.1 percent of the car drivers aged 18 - 30 years belong to this group but one third of the older ones. The highest part with heavy alcoholic problems is to find in the group of 41 - 50 years with 38,6 percent. The high cut-offs were exceeded at GGT in 70 cases, methanol in 30, CDT in 29, BAC in 28 and AcIp in 17. The part of car drivers with indicators proving alcohol abuse is significantly increased at blood ethanol levels higher than 2.00 g/kg. Below this concentration there are no significant differences between groups of BAC like 0.80 - 1.29 g/kg, 1.30 - 1.59 g/kg or 1.60 - 1.99 g/kg. Car drivers with indicators above these high cut-offs are to find also at low BAC's. Therefore BAC alone is no sufficient criterion to detect alcohol misuse.


Caspar, C B et al.: Schweiz. Med. Wschr. 123, 1929-1934 (1993)

Crato, H et al.: Beitr. Gerichtl. Med. 36, 275-279 (1978)

Iffland, R et al.: Beitr. Gerichtl. Med. 42, 231-236 (1984)

Iffland, R et al.: Blutalkohol 26, 87-97 (1989)

Iffland, R et al.: Blutalkohol 31, 273-314 (1994)

Iffland, R, Grassnack, F: Blutalkohol 32, 26-41 (1995)

Krambrich, T: Dissertation Köln (1993)

Majchrowicz, E, Mendelson, J H: J. Pharm. Exp. Therap. 179, 293-300 (1971)

Tiess, D: Z. ges. Hygiene 31, 530-531 (1985)

Tiess, D, Stöhlmacher, P: priv. communic. (1990)