Alcohol & Alcoholism Vol. 44, No. 1, pp.

93102, 2009 Advance Access publication 25 November 2008

doi: 10.1093/alcalc/agn095

EPIDEMIOLOGY The Composition of Alcohol Products from Markets in Lithuania and Hungary, and Potential Health Consequences: A Pilot Study
Dirk W. Lachenmeier1, , Bart Sarsh2 and J urgen Rehm2,3,4
1 Chemisches

und Veterin aruntersuchungsamt (CVUA) Karlsruhe, Germany, 2 Centre for Addiction and Mental Health (CAMH), Toronto, ON, Canada, 3 Institut f ur Klinische Psychologie und Psychotherapie, TU Dresden, Germany and 4 Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada Corresponding author: Chemisches und Veterin aruntersuchungsamt (CVUA) Karlsruhe, Weissenburger Strasse 3, D-76187 Karlsruhe, Germany. Tel: +49-721-926-5434; Fax: +49-721-926-5539; E-mail: Lachenmeier@web.de (Received 20 November 2007; first review notified 29 January 2008; in revised form 13 F ebruary and 31 July 2008; accepted 16 October 2008; advance access publication 25 November 2008) Abstract Aims: The rates of alcohol-attributable mortality in Lithuania and Hungary have been shown to be higher than those in most other European countries. Quality of alcohol products is investigated as a possible explanation. Methods: In a descriptive pilot study, a convenience sample of alcohol products was collected from local city markets in both countries (Lithuania n = 10, Hungary n = 15) and chemical analyses, including some that have not been done in prior studies, were conducted. The parameters studied were alcoholic strength, volatiles (methanol, acetaldehyde, higher alcohols), ethyl carbamate, anions (including nitrate) and inorganic elements (including lead). Additionally, a multi-target screening analysis for toxicologically relevant substances was conducted. Results: The majority of samples (64%) had an alcohol content between 35% vol. and 40% vol., being in accordance with the typical strength of legal spirits in Europe. Three samples containing significantly higher concentrations of alcohol above 60% vol. were found to be unrecorded alcohol products, defined as any alcohol that is outside of legal and taxed production. Screening analysis showed that those samples contained various flavourings, including the hepatotoxic substance coumarin, at concentrations above the legal limit for foods. All other substance classes under study were found to be at levels of no toxicological concern. Conclusions: Although some problems with the quality of the alcohol samples were found, there is insufficient evidence from this pilot study to conclude that alcohol quality has an influence on health as reflected in alcohol-attributable mortality rates. Given the extent of alcohol-attributable disease burden in central and eastern European countries, future research should focus on collection of large, representative samples, particularly of unrecorded sources, which was the most problematic product group in our study.

Downloaded from http://alcalc.oxfordjournals.org/ by guest on June 4, 2013

INTRODUCTION Alcohol is a major risk factor for burden of disease (Ezzati et al., 2002), with an especially high impact in central and eastern Europe (Rehm et al., 2003c, 2004). Rehm and colleagues have estimated the rate of alcohol-attributable mortality in several European countries for the year 2002. In both Hungary and Lithuania (Lituvania), the alcohol-attributable mortality rate for men was 22 per 10,000 population between the ages of 20 and 64, while other central and western European countries that were examined (e.g. Czech Republic, France, Poland, Sweden and the United Kingdom) had rates that were less than 10 per 10,000 (Rehm et al., 2007). Only Russia had a higher rate of alcohol-attributable mortality at 29 per 10,000 men. Female mortality was also elevated in Hungary, Lithuania and Russia, albeit at a lower level (between 4 and 5 per 10,000; with rates below 2 for the other countries examined). Both Hungary and Lithuania are among the top three EU countries with respect to alcohol-attributable mortality (Rehm, 2007). Volume of alcohol consumption alone cannot explain these differences, as both countries are within the range of other EU countries in per capita consumption, including unrecorded consumption (Popova et al., 2007). Although heavier drinking occasions in general have been found to exist in central and eastern European countries (e.g. Popova et al., 2007), drinking patterns in Lithuania and Hungary did not differ from neighbouring countries such as Poland, Slovakia or other Baltic countries. In fact, results from a recent comparative EU survey indicate that Hungary and Lithuania rank below the EU-25 average in rates of heavy drinking (Eurobarometer, 2007). Thus, drinking patterns are unlikely to explain the more than 2-fold

rates of alcohol-attributable mortality in Hungary and Lithuania compared to a country like Poland (Rehm et al., 2007). To account for the high rates of alcohol-attributable mortality, other dimensions of alcohol consumption in Hungary and Lithuania may be exerting a negative impact on health. One plausible explanation is alcohol quality. Based on chemical analyses of samples from Hungary, Sz ucs et al. (2005) found that consumption of homemade spirits may be a risk factor for the development of alcohol-induced cirrhosis. However, it would be premature to assert that this was the only reason for the very high rate of cirrhosis in Hungary. Researching spirits in Estonia, Lang et al. (2006) advanced two hypotheses on the relationship between alcohol quality and health. First, long-chain alcohols may have specific detrimental health consequences, including cirrhosis. Second, the high alcohol content of some spirits may be linked to acute consequences, especially alcohol poisoning. Similar arguments with respect to so-called surrogate alcohol have been made in other studies (e.g. see McKee et al., 2005; Stickley et al., 2007). There has been no consistent definition of surrogate alcohol in the literature. However, the WHO recently provided a nomenclature about alcohol products including recorded and unrecorded ones, which we used in this study (see Fig. 1 for details; see also the Global Information System on Alcohol and Health on www.who.int). Unrecorded alcohol products, according to this definition, include homemade alcohols, illegally produced or smuggled alcohol products as well as surrogate alcohol that is not officially intended for human consumption. Some common examples of surrogate alcohol include mouthwash, perfumes and eau-de-colognes, which are alcohol products manufactured on a large scale. Such alcohol

The Author 2008. Published by Oxford University Press on behalf of the Medical Council on Alcohol. All rights reserved

94

Lachenmeier et al.

Alcohol products

Note: Surrogate alcohol may be intended for human consumptionbut not declared as such, to evade taxes

Recorded consumption: taxed alcohol

Unrecorded consumption
30% globally

Examples:

Legal but unrecorded alcohol products (homemade or other)

Alcohol products recorded, but not in the jurisdiction where consumed Cross-border shopping; medicinal products for human intake

Surrogate alcohol: non-beverage alcohol products not officially intended for human consumption

Illegally produced or smuggled alcohol products intended for human consumption (including illegal homemade alcohol)

Downloaded from http://alcalc.oxfordjournals.org/ by guest on June 4, 2013

Commercial and taxed beer, wine, spirits

Homemade fruit spirits; homebrewed beer; wine products for home consumption

Cosmetics (mouth-wash, perfumes, etc); denatured alcohol; automobile products; medicinal compounds such as rubbing alcohol

Moonshine; samogon; untaxed beer, wine, or spirits

Fig. 1. Classification of alcohol products.

may be produced with human consumption in mind, but to evade taxation it is officially classified as shaving water or mouthwash. Consumption of unrecorded alcohol products is relatively prevalent in eastern Europe. Popova and colleagues (2007) estimate that >25% of alcohol consumed in Hungary and Lithuania is unrecorded, with a high level of surrogate alcohol expected to be comprised in these numbers, especially for Lithuania. Both of Lang et al.s (2006) hypotheses appear promising. Yet, studies that have examined these hypotheses also suffer from several methodological limitations. In fact, the Lang et al. (2006) study itself has two noteworthy limitations. First, the sample of legal spirits used for comparison with the surrogate samples appears to be biased towards products with relatively low concentration of long-chain alcohols such as vodka, brandy and whiskey but not fruit or marc spirits. Also, the current legal requirements for the composition of spirits in the European Union were overlooked (European Council, 1989; European Parliament and Council, 2008). Finally, only a very limited range of compounds (e.g. ethanol and higher alcohols) was analysed. Currently, there are no available data on healthrelevant substances like acetaldehyde, metals, anions and ethyl carbamate. The present research contributes to the literature on alcohol quality in Hungary and Lithuania, and its potential links to health outcomes. For the purposes of the pilot study, we operationalized alcohol products as including, in addition to legal alcoholic beverages, surrogate alcohol (see the definition of surrogate alcohol above; see also Lachenmeier et al., 2007). Samples of alcohol products were collected from local markets

within both countries and chemical analyses, including some that have not been done in prior studies, were conducted. This study was a pilot project examining the link between alcohol quality and health. METHODS Collection of samples Collection of alcohol products from Lithuania was based on a convenience sampling procedure. All samples (n = 10) were collected by means of cash transactions from a local market (i.e. a local flea or farmers market) in each city (Table 1). Products that would not ordinarily be sold in mainstream retail supermarkets can be purchased, often at highly discounted prices, at these markets. One of the markets (Kaunas) from which the samples were collected was located inside a building while the other (Utena) was an outdoor market (i.e. non-permanent market stalls were erected). In general, most of these types of markets in Lithuania sell alcohol products, including surrogate alcohol. A total of n = 7 samples were collected in Kaunas, which is a major metropolitan city, while the remaining samples were obtained from a smaller city (Utena). In each city, samples were collected by randomly selecting a bottle without the assistance of the market seller. Samples L1 and L5 through L10 were purchased from the same seller, and samples L2 through L4 were purchased from the same seller. At no point during sample collection did a seller imply that a surrogate bottle was being purchased. The only exception was with one of the eaude-cologne samples from Utena (sample L4). For that sample,

Composition of Alcohol
Vilijampole market, Kaunas, Lithuania Stumbro starka Trauktine (bitter alcohol)

95

the seller attempted to avoid the sale, claiming that the product was not very good. No further inquiry was made into the nature of the sample and the transaction proceeded. Regarding bottle packaging, only samples L5 through L10 contained a holographic sticker across the cap, presumably as a guarantee of authenticity. Moreover, apart from the three cosmetic samples (i.e. samples L2, L3 and L4), all samples had caps that could be opened without breaking the seal. No other peculiar aspects to the samples were recorded. With one exception (sample H15), the Hungarian samples were collected from three markets in the city of Budapest, including a large, open market with counterfeit goods locally known as the Tiger market (Table 2). All samples were obtained by cash. They all appeared to be legal alcoholic beverages, even though the prices in the Tiger market were much lower than all other sales points in the city. Samples were taken at random from the liqueur and spirits section of the stands. Analytical procedure Alcoholic strength was determined by FTIR spectroscopy according to the method described in Lachenmeier (2007). Volatile components were analysed on the basis of the European Community Reference Methods for the Analysis of Spirits using gas chromatography (GC) with a flame-ionization detector (FID) (European Commission, 2000). Additional details on the GC-FID procedure are published elsewhere (Lachenmeier et al., 2006b). Ethyl carbamate (urethane) was determined using GC with tandem mass spectrometry (GC-MS/MS) (Lachenmeier et al., 2005a). Anionic composition was analysed using ion chromatography (Lachenmeier et al., 2003). Conductivity was measured using the procedure outlined in Lachenmeier et al. (2008a). Inorganic elements were analysed using semiquantitative inductively coupled plasma MS (ICP-MS) after evaporation of the sample and re-constitution in ultrapure water. Furthermore, all samples were screened for unknown substances using high-performance liquid chromatography with a diode-array detector (HPLC-DAD) and GC-MS. Coumarin was quantified using HPLC-DAD (Sproll et al., 2008) and the results were confirmed using HPLC-MS. Indication of results Alcoholic strength is indicated as percent by volume (% vol.). Volatile compounds contained in each sample are expressed in the unit g/hl of pure alcohol or g/hl of 100% vol. alcohol (i.e. the concentrations are standardized in relation to alcoholic strength) according to the procedure described in the European Community Reference Methods for the Analysis of Spirits (European Commission, 2000). This approach is superior to reporting in mg/l because the samples can be directly compared irrespective of their individual alcoholic strength. For clarity, we use the abbreviation g/hl p.a.. Results for non-volatile components are presented as mg/l.

L10 L9 L8

Vilijampole market, Kaunas, Lithuania Kvietine degtine (brandy)

Vilijampole market, Kaunas, Lithuania Bajoru degtine (vodka)

Vilijampole market, Kaunas, Lithuania Factum brandy (brandy)

Downloaded from http://alcalc.oxfordjournals.org/ by guest on June 4, 2013

Table 1. Alcohol samples collected in Lithuania

L7 L6 L5

Utena, Lithuania Utena, Lithuania Utena, Lithuania

L4

L3

L2

L1

Vilijampole market, Kaunas, Lithuania Grand Cavalier (brandy)

Eau de cologne

Eau de cologne

Eau de cologne

Vilijampole market, Kaunas, Lithuania Palanga trauktine (bitter alcohol)

Vilijampole market, Kaunas, Lithuania Lithuanian Vodka Originali (vodka)

Sampling site

RESULTS
Description

Results for alcoholic strength and volatile composition are outlined in Table 3. The alcoholic strengths of the samples ranged from 35.7% vol. to 75.2% vol. (Lithuania), and between 29.9% vol. and 40.4% vol. (Hungary). The highest alcoholic

Sample

96

Lachenmeier et al.

Sampling site Central market Central market Central market Central market Lehal market Description Liqueur from Apricot Plum brandy Apricot Apricot apricots brandy Beregi brandy liqueur Barack Barack szilva Barack Barack szeszestal szeszestal szeszestal

strength was found in the three eau-de-colognes from Lithuania. These samples contained between 64.7% vol. and 75.2% vol. of alcohol. Methanol was detected in concentrations ranging from not detectable to 29 g/hl p.a. (Lithuania) and 20 g/hl p.a. to 616 g/hl p.a. (Hungary). Samples labelled as vodka (e.g. samples L6 and H8) had the lowest methanol content, while higher methanol levels were found in some of the Hungarian fruit-derived spirits (e.g. samples H1H2, H10H14). However, some samples that were labelled as fruit spirits also contained a low concentration of methanol (e.g. samples H3H7). Among the higher alcohols, content varied considerably between not detectable and 291 g/hl p.a. in the Lithuanian samples, and between not detectable and 1411 g/hl p.a. in the Hungarian samples. The same degree of variability was found for ethyl acetate, ethyl lactate and acetaldehyde. All elements that were positively detected during our ICP/MS screening analysis for elemental composition are shown in Table 4. The most abundant elements, with concentrations in the mg/l range, were the alkali and alkaline earth metals, sodium (4147 mg/l), potassium (1330 mg/l), calcium (117 mg/l) and magnesium (0.124 mg/l). Other metals were found in traces below 1 mg/l. The conductivities of the samples ranged from 2 S/cm to 479 S/cm (see Table 5). Nine of the samples were positive for chloride (2.253.1 mg/l) and sulphate (5.5507 mg/l), whereas only five samples contained nitrate (2.148.7 mg/l) and phosphate (18.051.6 mg/l). Ethyl carbamate was detected in three of the Hungarian stone-fruit spirits using GC-MS/MS (0.210.22 mg/l). During the multi-target screening analyses for unknown substances, which was done using HPLC-DAD and GC-MS, additional toxicologically relevant substances were discovered in the three cosmetically derived surrogates from Lithuania (samples L2 L4). The samples contained an abundance of flavour substances in the mg/l range (e.g. anethole, sabinene, camphene, pinene, coumarin and eugenol). From these, we quantified coumarin, which was found in concentrations between 16.7 mg/l and 33.0 mg/l.

H10, H11, H12 H9

Tiger market Liqueur

Tiger market Apricot brandy Barack parlat

Tiger market Apricot brandy Barack parlat Kecskemetl

H13, H14

Hotel Liqueur Keseru liguer

H15

Downloaded from http://alcalc.oxfordjournals.org/ by guest on June 4, 2013

Table 2. Alcohol samples collected in Hungary

H5

H6

H7

Lehal market Cherry liqueur Korte aromaval

Tiger market Vodka Alaska

H8

DISCUSSION Alcoholic strength The majority of samples (64%) had an alcohol content between 35% vol. and 40% vol., which is in accordance with the typical strength of legal spirits in Europe (Lachenmeier and Musshoff, 2004). Only three samples, which contained significantly higher concentrations of alcohol above 60% vol., were found to be surrogate samples. Our findings are similar to a study on Estonian surrogate alcohol by Lang et al. (2006). In Lang et al.s study, the mean concentration of illegally homemade alcohol (moonshine) was 42.8% vol., while cosmetic surrogates (e.g. aftershaves, etc.) contained an average of 64.2% vol. of alcohol. Compared to legal alcoholic beverages or homemade unrecorded alcohol, these high-strength cosmetic surrogates may have more pronounced detrimental health effects entirely explained by the presence of ethanol. Nonetheless, testing the extent to which ethanol affects health in countries of the former Soviet Union is complicated by the

Sample

H1, H2

H3

H4

Table 3. Volatile composition of Lithuanian and Hungarian alcohol in comparison to EU limits and data from the literature Lithuania g/hl p.a. L1 L2 L3 L4 L5 L6 L7 L8 L9 L10 H1 H2 H3 H4 H5 H6 H7 Hungary H8 H9 Comparison data H10 H11 H12 H13 H14 H15 EU maximum Samogon from Legally level for neutral Russia distilled alcohol/for fruit (Nuzhnyi, fruit spirits a spirits 2004) (Lachen(European meier and Parliament and Musshoff, Council, 2008) 2004)a No data 0164 1.3212 2142 065 053 4.7967 91170 No data No data 038 No data 0.8164 No data 066 No data No data 0250 No data 31.249.1 251389 No data 161393 155 nd368 15813 4456 No data No data No data No data No data No data No data No data No data No data No data

Ethanol (% vol.) Methanol Acetaldehyde 1-Propanol 1-Butanol 2-Butanol Isobutanol Amyl alcohols 1-Hexanol Benzyl alcohol 2-Phenyl ethanol Methyl acetate Ethyl acetate Benzyl acetate Ethyl lactate Ethyl caprylate Ethyl benzoate Benzaldehyde Sum of higher alcohols

37.2 75.2 66.4 64.7 36.8 39.8 37.5 39.8 35.7 42.6 29.9 30.2 34.9 37.3 37.5 30.5 30.9 37.1 30.2 40.0 40.0 40.0 40.0 40.0 40.4 29 21 23.7 1.6 0.8 87 174 0.6 nd 3.6 nd 23.8 nd 14.4 nd 6.1 nd 1.8 nd nd nd nd nd nd nd nd nd nd nd nd nd 4.5 2.2 nd nd nd nd nd nd nd nd nd nd nd nd nd nd 2 0.8 5.9 nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 2.5 5.5 0.5 nd nd 1.3 4.1 nd nd nd nd 2.3 nd nd nd nd nd 6 nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 14.9 0.6 nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 26.1 27.3 43.4 1.0 1.8 26.2 113 1.0 nd 2.0 nd 44.6 nd 61.2 nd 1.4 5.3 nd nd nd 5.3 9.2 nd nd nd nd nd nd nd nd 260 7.5 15.5 0.3 1.7 11.7 25.2 1.7 2.1 nd nd 18.4 0.3 2.9 0.3 464 19.4 32.6 1.1 3.4 23.5 48.8 4.7 5.6 nd 1.1 33.3 0.5 7.1 0.4 30 4.2 0.3 0.5 0.9 nd 0.6 nd 12.6 nd nd 3.8 nd nd nd 75 1.6 40.4 0.9 4.3 5.2 11.0 0.6 0.5 nd nd 36.9 nd nd nd 34 nd 0.3 nd nd nd 0.9 nd nd nd nd nd nd nd nd 52 1.4 9.5 nd 0.6 1.6 1.6 nd 3.1 nd nd 1.0 nd nd nd 33 0.5 0.8 nd nd 0.5 0.6 nd 7.2 nd nd 1.9 nd nd 0.3 20 nd nd nd nd nd nd nd nd nd nd nd nd nd nd 155 nd 0.6 nd nd 0.6 1.1 nd nd nd nd 0.9 nd nd nd 403 4.9 15.8 1.9 0.6 9.8 15.6 0.6 nd 0.3 nd 17.1 nd 0.7 0.5 355 3.8 13.6 1.6 0.6 8.8 14.0 0.5 nd 0.3 nd 13.8 nd 0.6 nd 616 4.4 998 22.1 130 58.1 192 11.1 0.8 1.2 6.0 159 nd 50.4 0.8 588 4.2 972 21.7 129 57.3 189 11.0 0.6 1.2 5.9 161 nd 46.1 0.7 592 4.3 994 22.1 131 58.3 193 11.2 0.7 1.3 5.9 159 0.3 49.0 0.8 45 4.7 0.5 nd nd nd nd nd nd nd nd 2.1 0.3 0.3 nd 30/1000 0.5/no limit 1.3/no limit

Composition of Alcohol

nd nd nd nd 291 2

nd nd nd nd 188 15

3.2 0.3 nd 0.3 nd nd nd nd 5.7 13.4 7.7 14.4 4.6 7.5 nd nd 56 114 2.3 62.4 1.2 13.3 1.9 nd

nd nd nd 0.5 0.5 0.5 0.3 nd 0.3 0.3 1.4 1.3 1.4 nd 2.3 44.3 39.1 1411 1380 1410 0.5 0.5/no limit (minimum 200)

nd, not detected (detection limit 0.5 g/hl p.a.). a Recalculated to g/hl p.a. under the assumption of an alcoholic strength of 40% vol.

97

Downloaded from http://alcalc.oxfordjournals.org/ by guest on June 4, 2013

98

Table 4. Inorganic composition of Lithuanian and Hungarian alcohol in comparison to WHO and EU limits (results of ICP-MS semiquantitative analysis) Lithuania (mg/l) L1 L2 L3 L4 L5 L6 L7 L8 L9 L10 H1 H2 H3 H4 H5 Hungary H6 H7 H8 H9 Comparison data H10 H11 H12 H13 H14 H15 EU drinking water WHO guidelines quality for drinking standards water quality (WHO, 2006) (European Council, 1998) nd 0.10 nd 1.3 nd 0.01 1.4 0.2 8.6 nd 0.02 nd nd 0.4 nd nd nd 0.05 nd nd 0.07 nd 1.2 nd 0.02 1.3 0.2 8.4 nd 0.02 nd nd 0.3 nd nd nd 0.17 nd nd 0.76a nd 3.6 nd 0.02 3.2 0.9 5.6 nd 0.16 nd nd 0.2 nd nd nd 0.09 nd nd 0.11 nd 4.1 nd 0.01 1.3 1.0 5.4 nd 0.07 nd nd 0.2 nd nd nd 0.18 nd nd 0.12 nd 3.9 nd 0.07 1.4 0.9 4.8 0.004 0.14 nd nd 0.1 nd nd nd 0.16 nd nd 0.16 0.006 17 0.004 0.08 282 24 147 0.015 7.2 0.004 0.17 8.6 nd 0.010 0.003 0.41 nd 0.2 1.0 2.0 0.02 0.01 0.005 0.5 2 0.07 0.01 0.02

Al B Bi Ca Co Cu K Mg Na Ni P Pb Rb Si Sb Ti V Zn Zr

0.44a 0.15 nd 5.4 nd 0.52 2.6 1.1 9.7 nd 0.2 0.006 0.0021 0.44 nd nd nd 0.17 nd

nd 0.03 nd 1.1 nd 0.01 1.3 1.4 7.6 nd nd nd nd 0.34 0.006a nd nd nd nd

nd 0.05 nd 1.9 nd nd 2.3 4.3 11 nd nd nd nd 0.47 0.003 nd nd nd nd

nd 0.13 nd 2.7 nd nd 1.0 0.2 5.9 nd nd nd nd 1.2 0.004 nd nd 0.07 nd

1.2a 0.55a nd 11 0.003 0.15 219 12 17 0.005 2 0.015a 0.25 2.5 nd 0.006 0.012 0.32 0.026

0.6a 0.08 nd 0.8 0.002 0.03 1.5 0.1 5.5 nd nd 0.008 nd 0.1 nd nd nd 0.10 0.010

nd 0.04 nd 0.4 nd 0.01 3.1 0.06 5.1 nd nd nd 0.0015 0.14 nd nd nd nd nd

nd 0.25 nd 0.3 nd nd 1.3 0.04 12 nd nd nd nd 0.18 nd nd nd nd nd

0.46a 0.45 nd 1.5 nd 0.82 1.6 0.2 4.3 0.002 nd 0.011a 0.0016 0.36 nd nd nd 0.13 nd

0.69a 0.45 nd 5.8 0.002 0.11 96 6.2 6.6 nd 1.2 0.011a 0.054 0.86 nd 0.005 0.004 0.14 0.006

0.15 0.75a 0.007 11 nd 0.07 330 9.6 44 0.010 4.8 0.009 0.11 1.9 nd 0.010 0.003 0.30 0.006

0.11 0.77a 0.003 11 nd 0.03 339 9.1 44 0.008 4.8 0.005 0.12 2.0 nd 0.010 0.003 0.26 0.009

0.02 0.04 nd 1.2 nd 0.02 1.4 0.2 80 nd 0.4 nd nd 1.7 nd nd nd nd nd

nd 0.06 nd 1.1 nd 0.11 1.1 0.2 114 nd 0.07 nd nd 2.1 nd nd nd nd nd

nd 0.08 nd 1.5 nd 0.01 203 0.3 12 nd 0.25 nd 0.014 5.3 nd 0.003 nd nd nd

0.01 0.05 nd 1.1 nd 0.05 1.4 0.2 106 nd 0.16 nd nd 2.9 nd nd nd nd nd

0.01 0.06 nd 1.0 nd 0.02 1.2 0.1 115 nd 0.09 nd nd 2.6 nd nd nd nd nd

nd 0.04 nd 0.6 nd 0.01 2.6 0.1 8 nd 0.79 nd nd 0.2 nd nd nd 0.06 nd

0.03 0.04 nd 1.1 nd 0.01 0.9 0.3 107 nd 0.03 nd nd 3.2 nd 0.002 nd nd nd

Lachenmeier et al.

nd, not detected. a Value above EU or WHO limit for drinking water (no limits for spirits available).

Downloaded from http://alcalc.oxfordjournals.org/ by guest on June 4, 2013

Composition of Alcohol
WHO/Codex Alimentarius standardsc

99

Table 5. Miscellaneous parameters (conductivity, anionic composition, ethyl carbamate, and coumarin) of Lithuanian and Hungarian alcohol in comparison to WHO and EU limits

H10 H11 H12 H13 H14 H15 EU standardsb

fact that surrogate alcohol is generally consumed by those with alcohol dependence who cannot afford other forms of alcohol (Stickley et al., 2007; Tomkins et al., 2007). Moreover, this population is expected to suffer from health problems as a result of consuming any alcohol, not necessarily high-strength surrogate alcohol. Thus, the effects of sample selection bias on estimates of attributable burden from different forms of alcohol such as surrogate alcohol are hard to control. Volatile composition Besides ethanol, our samples contained a number of volatile compounds, which are expected in products derived from alcoholic fermentation. Methanol is a substance most often associated with the toxicity of surrogate and other alcohols (Lachenmeier et al., 2007). The methanol content of the Lithuanian surrogates was relatively low (i.e. lower than the EU limit of 30 g/hl p.a. for neutral alcohol) (European Parliament and Council, 2008). The higher methanol content of the Hungarian samples indicates that they are manufactured from fruit materials. Release of methanol, which is bound in fruit pectines, typically leads to methanol concentrations above 100 g/hl p.a. The current EU limit for naturally occurring methanol in some fruit spirits is 1000 g/hl p.a. (which equates to 0.4% vol. methanol at 40% vol. alcohol). None of the samples exceeded this limit. The level above which toxic effects are expected (2% vol.) is substantially higher than the EU limit (Paine and Dayan, 2001). In our samples, it appears that methanol content did not pose a threat to public health. Acetaldehyde is an undesirable substance in spirits because of its unpleasant flavour. It is also regarded as possibly carcinogenic to humans [International Agency for Research on Cancer (IARC) Group 2B] (IARC, 1999). During distillation, acetaldehyde is enriched in the first fraction, which is generally discarded. During production of spirits, acetaldehyde may be formed not only as a product of alcoholic fermentation by Saccharomyces yeast, but also as a metabolite of microorganisms such as lactic acid or acetic acid bacteria (Pieper et al., 1987). An increased amount of acetaldehyde usually indicates flaws in the fermentation process. Using standard distillation stills, most of the acetaldehyde can be separated. Nevertheless, complete separation is not technically possible. The average acetaldehyde residue of international spirits is 17 25 g/hl p.a. (66 101 mg/l) (Lachenmeier and Sohnius, 2008). Apart from samples L1, L9 and H2, acetaldehyde levels were actually lower than this average found in legal spirits. Either the acetaldehyde fraction was cleanly separated during distillation or neutral alcohol, which only contains very low concentrations of acetaldehyde, was added to the fruit distillates. Alcohol containing more than two carbon atoms is commonly called higher or fusel alcohol. Most higher alcohols are created as a byproduct of yeast fermentation and are important flavour compounds. In legal alcoholic beverages, the content of higher alcohols is generally judged to be of no toxicological relevance. The Joint FAO/WHO Expert Committee on Food Additives has included higher alcohols (e.g. 1-propanol, 1-butanol and isobutanol) in the functional class flavouring agent and has found that there is no safety concern at current levels of intake when used as a flavouring agent (JECFA, 1997). For certain groups of spirits, the European Union even

Comparison data

50 250 50 250 53.1 11.7 51.6 84.0 nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 0.21 nd nd nd nd 0.22

2500

241

22

23

23

nd

nd

nd

nd

nd

nd

10

10

Downloaded from http://alcalc.oxfordjournals.org/ by guest on June 4, 2013

174 8

H8 H9

2.3 nd nd nd nd nd nd nd nd nd

Hungary

167 170 8

nd nd, not detected (detection limits: chloride 2 mg/l, nitrate 5 mg/l, phosphate 10 mg/l, sulphate 5 mg/l). a Value exceeding international standards. b EU drinking water standards (European Council, 1998); EU coumarin limit (European Council, 1988). c WHO guidelines for drinking water quality (WHO, 2006); Codex Alimentarius coumarin limit (Codex Alimentarius, 1985). nd 27.3a 16.7a 33.0a nd nd nd nd nd nd nd nd nd nd nd nd nd

H7

H6

H5

H3

138 169

H4

H2

463

L10 H1

479

131

L6 L7 L8 L9

10 145

Lithuania

L4

272

L1 L2

22

Conductivity (S/cm) Chloride (mg/l) Nitrate (mg/l) Phosphate (mg/l) Sulphate (mg/l) Ethyl carbamate (mg/l) Coumarin (mg/l)

36

2.9 nd nd 6.6 nd

nd nd nd nd nd

2.3 nd nd nd nd

L3

nd nd nd nd nd

7.2 nd 46.2 507a nd

L5

nd nd nd 5.5 nd

13

nd nd nd nd nd

nd nd nd nd nd

nd nd nd 31.9 nd

2.2 2.1 18.0 60 nd

10.8 48.7 48.7 143 nd

8.5 45.2 45.2 146 nd

nd nd nd nd nd

nd nd nd nd 0.22

16.3 6.6 nd 50.3 nd

244

nd nd nd nd nd

nd nd nd nd nd

nd

100

Lachenmeier et al.

demands a minimum volatile substance content (i.e. the quantity of volatile substances other than ethanol and methanol, which are mainly higher alcohols). For example, fruit spirits must have a volatile substance content of at least 200 g/hl p.a., whereas neutral alcohol should be almost entirely free of higher alcohols (max. 0.5 g/hl p.a.) (European Parliament and Council, 2008). The toxicity of higher alcohols was recently reviewed by Lachenmeier et al. (2008b). On the basis of no observed adverse effect levels (NOAELs) in animal experiments, acceptable daily intakes (ADI) and maximum limits in foods were extrapolated using a safety factor of 100. Based on the assumption that an adult consumes 4 25 ml of a drink containing 40% vol. alcohol, a reasonable preliminary guideline level would be 1000 g/hl of pure alcohol for the sum of all higher alcohols. Higher alcohols that were found in our samples were generally lower than the EU minimum requirement of 200 g/hl p.a.; only samples H12H14 contained unusually high concentrations above the level of 1000 g/hl p.a. (1380, 1410 and 1411 g/hl p.a., respectively). Only those three samples were a cause for concern, given the increased toxicity resulting from higher alcohols. However, due to limitations in the extrapolation of toxicity data from animals to humans, this association is far from unequivocally established. In general, the concentration of higher alcohols in our samples was lower than that reported in Russian samogon or legal fruit spirits (see Table 3 for a comparison). The relatively low concentration of volatile substances in the Hungarian fruitderived spirits, which was below the EU minimum requirement, is an indication that those products (samples H1H7 and H10 H11) have been adulterated with neutral alcohol and are not pure fruit distillates. This is suggestive of practices intended to deceive the consumer rather than being a threat to public health.

sorption spectrometry would be required to confirm the results of ICP-MS prior to official complaints. Applicable standards dealing with contaminants and toxins in foods require that contaminant levels remain as low as reasonably possible, and that contamination be reduced by applying appropriate technology in food production, handling, storage, processing and packaging (Codex Alimentarius, 1997). In this regard, we think that inorganic contaminants should be reduced in the spirits, especially in the drinking water from which they are derived. The samples were analysed for inorganic anions because of nitrate contamination, which may pose a health threat. Nitrate concentration was either not detectable or relatively low in all samples. The conductivities of the samples and the contents of the other anions were also relatively low, which suggests an overall sufficient level of water quality or water treatment. The drinking water limit was exceeded in sample L5 because of elevated sulphate levels. Ethyl carbamate Ethyl carbamate (urethane) may be formed naturally as a result of fermentation. It has been detected in a variety of fermented foods and beverages. Ethyl carbamate is probably carcinogenic to humans (IARC Group 2A) (IARC, 2007). The concentrations in wine and beer are usually below 100 g/l, while higher levels (in the mg/l range) have been found in some spirits. The highest ethyl carbamate concentrations were found in spirits derived from stone fruit of the species Prunus L. (Rosaceae) [e.g. cherries, plums, mirabelles (yellow plums) or apricots] (Battaglia et al., 1990; Zimmerli and Schlatter, 1991; Lachenmeier et al., 2005b). Canada, for example, has established an upper limit of 0.4 mg/l ethyl carbamate for fruit spirits (Conacher and Page, 1986). While ethyl carbamate was detected in three of the Hungarian fruit spirits, the concentrations were below this upper limit. Flavourings and coumarin All three Lithuanian cosmetically derived surrogates (e.g. aftershaves or perfumes that are consumed by individuals) contained different flavour compounds. Topical use of such products is generally regarded as harmless, but ingesting them may lead to health problems. The Codex Alimentarius (1985) provides general requirements for natural flavourings and has been introduced into European law (European Council, 1988). For some biologically active substances, maximum levels are specified. The most familiar example of these substances is thujone, which was made responsible for so-called absinthism (Lachenmeier et al., 2004, 2006a). However, thujone was not detected in our samples, although we detected coumarin in the three cosmetically derived samples during our GC-MS screening, which was confirmed by quantitative HPLC-DAD and HPLC-MS analysis. In all three samples, the coumarin concentration was above the Codex Alimentarius maximum level. Under European law, all three products would not be marketable if they were sold as food. In the 1980s, researchers hypothesized that coumarin has genotoxic and carcinogenic effects (AFC, 2004). More recent evidence, however, indicates that coumarin is not a genotoxic agent (Lake, 1999). The IARC has classified coumarin into group 3 (not classifiable as to its carcinogenicity in humans).

Downloaded from http://alcalc.oxfordjournals.org/ by guest on June 4, 2013

Non-volatile compounds and water quality Because elements and ions are generally non-volatiles, most of the inorganic content found in spirits is derived from dilution water used for the adjustment of the distillate to drinking strength. Inorganic contamination may occur from the use of the distillation equipment. For example, during the production of moonshine in the USA, the leaching of lead from solder or other lead-containing materials in automobile radiators used in the distillation stills caused lead contamination of the moonshine (Lachenmeier et al., 2007). In the Lithuanian and Hungarian samples, lead was detected in single cases. Three Lithuanian samples had lead concentrations above the drinking water standards of the WHO and EU (0.01 mg/l) (European Council, 1998; WHO, 2006). The drinking water standards were also exceeded by five Lithuanian samples for aluminium, one Lithuanian and three Hungarian samples for boron and one Lithuanian sample for antimony. For toxicity evaluation, it should be noted that the WHO and EU standards for metals are only valid for drinking water, which has a much higher daily rate of consumption than spirits. While some of the spirits under investigation slightly exceeded the drinking water limits, it cannot be concluded that such spirits are toxic or unsuitable for consumption. The screening nature of our assay for metals should also be noted, since reference analytics like atomic ab-

Composition of Alcohol

101

There is no epidemiological data on the carcinogenicity of coumarin and there is only limited evidence on the carcinogenicity of coumarin in experimental animal studies (IARC, 2000). Based on the NOAEL for hepatotoxicity, the Scientific Panel on Food Additives, Flavourings, Processing Aids and Materials in Contact with Food (AFC) has established a tolerable daily intake (TDI) of 0.1 mg/kg bw (AFC, 2004). Thus, a 60 kg adult would need to consume 182 ml of sample L4 to exceed the TDI of coumarin. A person with alcohol dependence is likely to exceed the TDI level. CONCLUSIONS Our research contributes to the literature on the relationship between alcohol quality and health because it quantified and analysed alcohol composition through chemical analyses of alcohol products. Overall, the quality of the samples was relatively high. In particular, except for ethanol, no compounds that may lead to acute toxic effects were detected. Although some problems with the quality of the samples were found, there is insufficient evidence from our pilot study to conclude that alcohol quality influences alcohol-attributable mortality rates. There are several methodological limitations to our study. For instance, only a limited number of samples were analysed. Furthermore, because of the convenience sampling strategy, our samples provide only a cross-section of the situation in one large and one smaller city in Lithuania, and the capital of Hungary. Given the small sample size, it might have excluded problematic alcohol products that were widely consumed. Indeed, if problems of alcohol quality in Lithuania and Hungary were linked primarily to home production, our results might be uninformative, since the effects of this factor on alcohol-attributable mortality were not addressed in our study. Similarly, there may be differences in the quality of alcohol in more rural areas of both countries. Nonetheless, the rates of alcohol-attributable mortality in both Lithuania and Hungary are more than double the rates of other European countries. Thus, one would expect to find some effects in the cities as well. Research on alcohol quality is important because it may ultimately provide an explanation for certain country-specific disease patterns, such as the high rate of liver cirrhosis in Hungary. Moreover, this study has demonstrated quality problems involving hepatotoxicity in the Lithuanian surrogate samples. This finding cannot account for the levels of alcohol poisoning and other acute effects of alcohol characteristics in Lithuania, but it may provide a link to the overmortality of chronic disease. In light of these considerations, we emphasize that the reported study is a pilot project that will serve as a foundation for future research, including one on unrecorded consumption in different parts of the European Union. Future research should focus on the collection of large, representative alcohol samples from both countries, including rural and urban regions. Given the extent of alcohol-attributable disease burden in central and eastern European countries, it is surprising that so little research has been conducted on the relationship between health and alcohol quality. Thus, apart from alcohol quality, more research on sources and patterns of consumption is needed. Particular focus should be placed on unrecorded sources (e.g. surrogate alcohol that is manufactured on a wide scale and homemade alcohol), which have been shown

to be relatively high in this region (Rehm et al., 2003a). Finally, casecontrol analyses for the most prevalent chronic and acute alcohol-related disease groups (Rehm et al., 2003b), including chemical analyses of alcohol products, should be conducted.
Acknowledgements The laboratory analyses were supervised by K. Hermann (ICP/MS), E.-M. Sohnius (GC-FID), T. Kuballa (GC-MS and GC-MS/MS), R. Attig (IC), M. Kohl-Himmelseher (HPLC) and C. Sproll (Coumarin analysis). The authors thank un and G. Bippes H. Heger, M. Jaworski, I. Hundeck, I. K ubel, H. Havel, K. M uller, M. Ur for excellent technical assistance.

REFERENCES
AFC. (2004) Opinion of the scientific panel on food additives, flavourings, processing aids and materials in contact with food (AFC) on a request from the commission related to Coumarin. EFSA J 104:1 36. Battaglia R, Conacher HBS, Page BD. (1990) Ethyl carbamate (urethane) in alcoholic beverages and foods: A review. Food Addit Contam 7:47796. Codex Alimentarius. (1997) Codex general standard for contaminants and toxins in foods (CODEX STAN 193-1995, Rev. 1-1997). www.codexalimentarius.net (31 July 2008, date last accessed). Codex Alimentarius. (1985) General requirements for natural flavourings (CAC/GL 29.1987). www.codexalimentarius.net (31 July 2008, date last accessed). Conacher HBS, Page BD. (1986) Ethyl carbamate in alcoholic beverages: a Canadian case history. In: Proceedings of Euro Food Tox II, 23742. European Society of Toxicology, Schwerzenbach, Switzerland. Eurobarometer. (2007) Attitudes towards alcohol. http://ec.europa.eu/ health/ph_determinants/life_style/alcohol/documents/ebs272_en. pdf (31 July 2008, date last accessed). European Commission. (2000) Commission Regulation (EC) no. 2870/2000 laying down community reference methods for the analysis of spirits drinks. Official J Eur Communities L333:2046. European Council. (1988) Council Directive (EEC) no. 88/388 on the approximation of the laws of the Member States relating to flavourings for use in foodstuffs and to source materials for their production. Official J Eur Communities L184:616. European Council. (1989) Council Regulation (EEC) no. 1576/89 laying down general rules on the definition, description and presentation of spirit drinks. Official J Eur Communities L160: 117. European Council. (1998) Council Directive 98/83/EC on the quality of water intended for human consumption. Official J Eur Communities L330:3254. European Parliament and Council. (2008) Regulation (EC) no. 110/ 2008 of the European Parliament and of the Council of 15 January 2008 on the definition, description, presentation, labelling and the protection of geographical indications of spirit drinks and repealing Council Regulation (EEC) no. 1576/89. Official J Eur Union L39:1654. Ezzati M, Lopez AD, Rodgers A et al. Comparative Risk Assessment Collaborating Group. (2002) Selected major risk factors and global and regional burden of disease. Lancet 360:134760. IARC. (1999) Acetaldehyde. In IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 71, Re-evaluation of Some Organic Chemicals, Hydrazine and Hydrogen Peroxide. Lyon, France: International Agency for Research on Cancer. IARC. (2000) Coumarin. In IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 77, Some Industrial Chemicals. Lyon, France: International Agency for Research on Cancer. IARC. (2007) IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 96, Alcoholic Beverage Consumption and Ethyl Carbamate (Urethane). Lyon, France, in press. JECFA. (1997) Summary of Evaluations Performed by the Joint FAO/WHO Expert Committee on Food Additives. Geneva: World Health Organization. Lachenmeier DW. (2007) Rapid quality control of spirit drinks and beer using multivariate data analysis of Fourier transform infrared spectra. Food Chem 101:82532.

Downloaded from http://alcalc.oxfordjournals.org/ by guest on June 4, 2013

102

Lachenmeier et al. Paine AJ, Dayan AD. (2001) Defining a tolerable concentration of methanol in alcoholic drinks. Hum Exp Toxicol 20:563 68. Pieper HJ, Rau T, Eller T et al. (1987) A speedy method to determine acetaldehyde, with particular consideration being given to quality inspection in the manufacture of fruit spirits. Deutsche Lebensmittel-Rundsch 83:351. Popova S, Rehm J, Patra J et al. (2007) Comparing alcohol consumption in central and eastern Europe to other European countries. Alcohol Alcohol 42:46573. Rehm J. (2007) Alcohol. In Zatonski, W (ed.). Closing the Health Gap in the European Union, Final Report of the HEM Project (Project no 2003121). Warsaw, Poland: The Maria SklodowskaCurie Memorial, Cancer Center and Institute of Oncology. Rehm J, Rehn N, Room R et al. (2003a) The global distribution of average volume of alcohol consumption and patterns of drinking. Eur Addict Res 9:14756. Rehm J, Room R, Graham K et al. (2003b) The relationship of average volume of alcohol consumption and patterns of drinking to burden of disease: an overview. Addiction 98:120928. Rehm J, Room R, Monteiro M et al. (2004) Alcohol use. In Ezzati M et al. (eds). Comparative Quantification of Health Risks. Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Vol. 1. 9591108. Geneva: World Health Organization. Rehm J, Room R, Monteiro M et al. (2003c) Alcohol as a risk factor for global burden of disease. Eur Addict Res 9:15764. Rehm J, Sulkowska U, Manczuk M et al. (2007) Alcohol accounts for a high proportion of premature mortality in central and eastern Europe. Int J Epidemiol 36:45867. Sproll C, Ruge W, Andlauer C et al. (2008) HPLC analysis and safety assessment of coumarin in foods. Food Chem 109:462 69. Stickley A, Leinsalu M, Andreev E et al. (2007) Alcohol poisoning in Russia and the countries in the European part of the former Soviet Union, 1970-2002. Eur J Public Health 17:44449. Sz ucs S, S arv ary A, McKee M et al. (2005) Could the high level of cirrhosis in central and eastern Europe be due partly to the quality of alcohol consumed? An exploratory investigation. Addiction 100:53642. Tomkins S, Saburova L, Kiryanov N et al. (2007) Prevalence and socioeconomic distribution of hazardous patterns of alcohol drinking: study of alcohol consumption in men aged 25-54 years in Izhevsk, Russia. Addiction 102:54453. WHO. (2006) Guidelines for Drinking-Water Quality. Geneva, Switzerland: World Health Organization. Zimmerli B, Schlatter J. (1991) Ethyl carbamate: analytical methodology, occurrence, formation, biological activity and risk assessment. Mutat Res 259:32550.

Lachenmeier DW, Attig R, Frank W et al. (2003) The use of ion chromatography to detect adulteration of vodka and rum. Eur Food Res Technol 218:10510. Lachenmeier DW, Emmert J, Kuballa T et al. (2006a) Thujonecause of absinthism? Forensic Sci Int 158:18. Lachenmeier DW, Frank W, Athanasakis C et al. (2004) Absinthe, a spirit drinkits history and future from a toxicological-analytical and food regulatory point of view. Deutsche Lebensmittel-Rundsch 100:11729. Lachenmeier DW, Frank W, Kuballa T. (2005a) Application of tandem mass spectrometry combined with gas chromatography to the routine analysis of ethyl carbamate in stone-fruit spirits. Rapid Commun Mass Spectrom 19:10812. Lachenmeier DW, Musshoff F. (2004) Volatile congeners in alcoholic beverages. Retrospective trends, batch comparisons and current concentration ranges. Rechtsmedizin 14:45462. Lachenmeier DW, Rehm J, Gmel G. (2007) Surrogate alcohol: what do we know and where do we go? Alcohol: Clin Exp Res 31:161324. Lachenmeier DW, Schehl B, Kuballa T et al. (2005b) Retrospective trends and current status of ethyl carbamate in German stone-fruit spirits. Food Addit Contam 22:3975. Lachenmeier DW, Schmidt B, Bretschneider T. (2008a) Rapid and mobile brand authentication of vodka using conductivity measurement. Microchimica Acta 160:28389. Lachenmeier DW, Sohnius E-M. (2008) The role of acetaldehyde outside ethanol metabolism in the carcinogenicity of alcoholic beverages: Evidence from a large chemical survey. Food Chem Toxicol 46:290311. Lachenmeier DW, Sohnius E-M, Attig R et al. (2006b) Quantification of selected volatile constituents and anions in Mexican Agave spirits (Tequila, Mezcal, Sotol, Bacanora). J Agric Food Chem 54:3911 15. Lachenmeier DW, Haupt S, Schulz K. (2008b) Defining maximum levels of higher alcohols in alcoholic beverages and surrogate alcohol products. Regul Toxicol Pharmacol 50:31321. Lake BG. (1999) Coumarin metabolism, toxicity and carcinogenicity: relevance for human risk assessment. Food Chem Toxicol 37:423 53. Lang K, Vali M, Sz ucs S et al. (2006) The composition of surrogate and illegal alcohol products in Estonia. Alcohol Alcohol 41:44650. McKee M, Suzcs S, Sarvary A et al. (2005) The composition of surrogate alcohols consumed in Russia. Alcohol: Clin Exp Res 29:1884 88. Nuzhnyi V. (2004) Chemical composition, toxic, and organoleptic properties of noncommercial alcohol samples. In Haworth A, Simpson R (eds). Moonshine Markets. Issues in Unrecorded Alcohol Beverage Production and Consumption. New York: BrunnerRoutledge, 17799.

Downloaded from http://alcalc.oxfordjournals.org/ by guest on June 4, 2013