EPR/81/16
August 1981
| FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS | ESN:FAO/WHO/UNU EPR/81/16 August 1981 |
| WORLD HEALTH ORGANIZATION | |
| THE UNITED NATIONS UNIVERSITY |
Item 2.3.2 of the Provisional Agenda
Joint FAO/WHO/UNU Expert Consultation on Energy and Protein Requirements
Rome, 5 to 17 October 1981
RECENT DATA ON ETHANOL CONTRIBUTION TO ENERGY REQUIREMENTS OF NORMAL INDIVIDUALS. UPPER LIMIT OF ALCOHOL INTAKE
by
R. Nordmann and G. Péquignot Université René Descartes, Paris and INSERM, Paris
1. CONTRIBUTION OF ETHANOL TO ENERGY REQUIREMENT
(R. Nordmann)*
As early as 1953, the Joint FAO/WHO Expert Committee recommended that “since the energy provided by ethanol is available for metabolic purposes, it should be considered in estimating the energy value of diets” (1).
If alcohol is a source of energy, then a person with balanced caloric intake who drinks alcohol will then have excessive caloric intake.
Roe (2) has summarized the circumstances under which alcoholic consumption can lead to excessive caloric intake as follows:
Frequent attendance at gatherings or receptions where alcoholic drinks and snack foods are served together;
In countries or regions where the wine drinking during or between meals is part of the cultural norm;
When excessive drinkers find themselves in conditions under which they drink or eat frequently throughout the day, particularly during work;
During experiments in which volunteers add alcoholic drinks to their regular diet.
*Professor, Department of Medicine, Paris-Ouest (Université René Descartes)
Belfrage et al. (3) administered as light beer a daily dose of 639 of alcohol (the approximate equivalent of 0.9 g/kg of body weight and 16 percent of the daily caloric intake) over a five week period to eight apparently healthy volunteers. Daily intake was divided into four doses, so that presumably blood alcohol levels never exceeded 0.30 g/l and dropped to nil two hours after each of the four drinks. With one exception, all experimental subjects reported a weight gain of 1–4 kg by the end of the experiment, corresponding to an increase of about 17 percent of the total caloric intake from alcohol. The authors also noticed slight but significant increases of plasmatic triglycerides during the first week of the experiment and plasmatic α-lipoproteins the fifth week. But there were no significant variations in cholesterol and plasmatic β-lipoprotein levels. A fleeting, but significant increase in hepatic triglycerides was also observed in three of the experimental subjects. This peaked during the third week.
The study by Belfrage et al. (3) is doubly interesting in that it shows how the addition of about 0.9 g of alcohol per kg of body weight does actually bring about a weight increase. This apparently “moderate” dose is nonetheless enough to induce perturbations in certain plasmatic and hepatic lipid parameters, at least in some experimental subjects.
Pirola and Lieber (4) did a study involving the daily administration of much larger doses of alcohol to habitual excessive drinkers. The experiment was conducted while the test subjects were in hospital and had not received alcoholic drinks or any kind of medication for at least six weeks. First, each subject was put on a controlled diet so that his body weight remained stable for at least one week. Next, part of the carbohydrate intake of 12 of these subjects was gradually replaced by ethanol. Based on a theoretical value of 7.1 kcal/g, ethanol accounted for 50 percent of the calories instead from day 9 to day 16. All other components of the diet except carbohydrates were modified. Under these conditions, Pirola and Lieber observed a weight loss of about 0.9 kg after 16 days of administration. This is a highly significant drop (p < 0.001 with respect to the control period which corresponds to a weight gain of 0.04 kg). The assumption that this weight loss might be due to caloric loss (of ethanol or some other nutrient) through elimination of urine or fecal wastes, was ruled out, when the authors measured the caloric content of these wastes with a calorimeter and found no variation in caloric content due to the introduction of ethanol into the diet (5).
The same authors studied the effect on weight of the addition of 2 000 kcal/day to the diet as ethanol to a basic daily diet of 2 500 kcal (7.1 kcal/g ethanol) (4). One subject had an initial weight gain of 1.1 kg, but his weight then dropped and was only + 0.190 kg on day thirty with respect to starting weight. So the average daily weight gain over the thirty days was only 6 g. The second subject was administered ethanol for only 11 days. The weight gain was 41 g/day for 11 days. Several weeks later, the same individual received an additional calorie increase of the same theoretical value (2 000 kcal/day), supplied as chocolate. He began to gain regularly; the final figure being 2.780 kg in two weeks, or about 198 g/day.
Pirola and Lieber findings therefore show that, under certain conditions, the true caloric value of ethanol for man is much below the theoretical figure of 7.1 kcal/g. It might be commented that these findings parallel those for animals. For example, the growth response of rats pair-fed in a theoretically isocaloric diet is markedly slower with the substitution of part of the carbohydrate intake (36 percent of the caloric intake) by ethanol (6,7).
Recent progress in the understanding of ethanol metabolism makes it possible to postulate several hypotheses to account for the different caloric values for ethanol found under different experimental conditions.
Ethanol is known to follow the metabolic pathway of alcohol dehydrogenase (ADH) as well as secondary pathways, primarily the microsomal ethanol-oxidizing system (MEOS) and catalase (8). Such secondary pathways come into play principally at high blood alcohol levels: thus the Km is 8.6 mM for MEOS, but only 0.5–2 mM, for ADH, corresponding to blood alcohol levels of 0.40 and 0.02 – 0.09 g/l respectively (4).
In the Belfrage et al. study (3) (administration to moderate drinkers of moderate doses of ethanol as alcohol drinks several times a day leading to blood alcohol levels below 0.30 g/l.) The ethanol was in all likelihood metabolized exclusively by the ADH pathway. ADH-induced ethanol oxidation is known to go hand-in-hand with the partial reduction of cytosolic NAD+ into NADH, which (through a shuttle system enabling cytosol reducer equivalents to be transferred into the mitochondria) supplies the respiratory chain enabling ATP synthesis to occur. Under such conditions, ethanol can be considered a source of energy available to the body with a theoretical caloric value equal or close to 7.1 kcal/g.
Pirola and Lieber's experiments (4,5), on the contrary, relate to intake of massive doses of ethanol by habitual excessive drinkers: for instance, the supplementary daily intake of 2 000 kcal in the form of ethanol by one individual (under conditions that the authors fail to specify) represents the equivalent of over 3 litres of 12° wine/day. Under the circumstances, blood alcohol levels must have been quite high for an extended period, so that a relatively large part of the ethanol might have been metabolized through secondary pathways (all the more so as MEOS is triggered by alcohol itself and that the individuals concerned were “alcoholics”). MEOS-induced ethanol metabolism results in the loss of chemical energy from both the substrate and the coenzyme (NADPH) without any known effective coupling to ATP synthesis. Presumably the energy unusable for ATP synthesis dissipates as heat (generated in excess of body's thermoregulatory needs) (5).
According to Lieber et al. (4,5,7), the failure to utilize ethanol calories observed in repeated ingestions of large quantities of alcohol would be explained by the action of the MEOS with its striking characteristics (relatively high Km; ability to be induced by ethanol).
But other mechanisms may be responsible for calorie loss under such conditions. Quite a few recent works have found the rate of ethanol oxidation in the liver to be indirectly regulated by the rate of ATP utilization. Increased ATP utilization thus has the effect of stimulating the re-oxidation of NADH by the respiratory chain, thus increasing ethanol oxidation by the ADH pathway (disassociation of the ADN-NADH compound being the limiting stage in ADH intervention) (9). Repeated ingestions of ethanol create a “hypermetabolic” state manifested in the liver by increased oxygen consumption and increased ethanol oxidation (10). This hypermetabolic state (not unanimously acknowledged (11)) is thought to be linked to stimulated ATP-consumption by cells (the exact nature of which is still controversial (12)). This condition would lead to calorie loss by a mechanism bypassing MEOS intervention.
Whatever the exact mechanism operating to produce the low ethanol calorie value associated with repeated, excessive drinking, the basic conclusion of the experiments on volunteers on controlled diets is that the contribution of ethnol to energy requirements depends basically on the conditions under which alcoholic drinks are administered.
Here it should be noted that, countrary to former belief, the rate of ethanol oxidation is not independent of the dose administered or ingested. It has in fact been ascertained for both rats (13,14) and man (14), that ethanol oxidizes faster when blood alcohol levels are significantly raised. Feinman et al. (14), studying in 12 male subjects, the rate of ethanol dissipation at alcohol blood levels of about 0.80 – 1.40 g/l and 0.18 – 0.80 g/l, showed that such rate is about 20 percent higher (p < 0.001) at blood alcohol levels exceeding 0.80 g/l. The relationship between the speed of ethanol oxidation and blood alcohol levels is presumably the result of secondary pathways of ethanol metabolism. Since we have already made the point that the utilization of these secondary pathways results in poor utilization of ethanol as a source of energy to fullfil requirements, it appears obvious that the same quantity of alcohol would better satisfy these requirements if taken under conditions producing only low or moderate alcohol blood levels. But it seems impossible to establish a precise correlation between caloric values of ethanol and alcohol blood levels, because the rate of ethanol oxidation does not depend exclusively on alcohol blood levels, but, as emphasized before, may be induced by secondary pathways or hypermetabolic states in “alcoholic” test participants. Eriksson and Peachy (15) observed that the rate of ethanol elimination after ingestion of a dose of 1 g/kg was 0.108 ± 0.012 g/kg/h in ten test participants who were not habitual drinkers and 0.120 ± 0.015 g/kg/h in ten “alcoholic” test participants.
Ethanol clearance further depends on nutritional factors. The earlier FAO report (1) focused on accelerated ethanol oxidation in carbohydrate - and protein-rich diets and slowed-down ethanol oxidation in high-fat diets and during fasting. Recent works have confirmed these views in greater detail.
Bode (16), for instance, observed that blood ethanol clearance in humans slowed by roughly 42% after a fast of 36 hours. Lumeng et al. (17) have established that reduced ethanol oxidation in rats under the influence of fasting is basically due to a drop in concentrations and/ or activities of ADH.
Concerning the increased rate of ethanol of ethanol oxidation induced by fructose, it is now assumed that this effect is linked to the increased ATP consumption required for fructose phosphorylation. This would stimulate the respiratory chain and as a result stimulate the re-oxidation of the NADH produced in ethanol oxidation (18). The practical applications of the fructose effect are very limited, as it only significantly stimulates alcohol metabolism when very large doses are administered, preferably intravenously (19). We should add that Clark et al. (20), assert that sorbitol, and to a smaller extent galactose, act on humans in the same way as fructose. But the same cannot be said of glucose (19).
Concerning the effect of protein intake in the diet, Bode et al. (19) showed that ethnol oxidation slowed by about 30 – 50 percent when administered for one week to test participants on isocaloric but protein poor diets. When a more physiological protein intake was re-established, ethanol oxidation returned to normal levels in a few days. We ourselves made a comparative survey of the effect of various natural amino acids on alcohol oxidation in isolated hepatocytes in rats (21). While certain amino acids do notably activate oxidation, especially in isolated hepatocytes in fasting animals, it would be easier to accelerate metabolism in humans by providing a protein-rich diet rather than by administering amino acids (22).
While the nature of the diet strongly influences the rate of ethanol oxidation, muscular exercise apparently affects oxidation much less. The earlier FAO report (1). stated that muscular exercise has no effect on the speed of alcohol oxidation. This conclusion was based partiuclarly on Pawan's work (23) on the effects of alcohol levels in blood and urine after running three miles or swimming 1 000 yards (volunteers receiving 0.5 g/kg of ethanol). Sautier et al. (24) have also determined that 480 kJ of intermittent muscular exercise in 4 hours (49 kms at 18 km/hour on a stationary bike) did not affect blood alcohol levels in volunteers having ingested 0.35 or 0.50 g of ethanol (marked with 13C) per kg or body weight. But the measurement of the expired 13 CO2 showed a very significant rise in the amount of ethnol fully oxidized into CO2.
The facts we have just related obviously indicate that ethanol's contribution to the energy requirement depends on a number of variables, in particular, the conditions under which alcoholic drinks are administered, the nutritional characteristics of the diet and muscular exercise, not to mention genetic factors which also influence ethanol metabolism. So it is not surprising, under these conditions, that it has not been possible to establish an unshakable relationship between alcohol consumption and obesity (25).
Studies in the USSR have shown a higher prevalence of obesity in the eastern part of Georgia (a wine-producing area) than in the western part, where alcohol consumption is lower (26). Although the authors were careful to compare only subjects belonging to similar rural populations with similar physical activities, their findings do not go beyond a mere presumption of a relationship between body weight and alcohol consumption.
In an American study with over 100 000 subjects, Klatsky et al. (27) observed a slight positive relationship between alcohol consumption and the Quetelet index (weight/height2 × 100) (used as an adiposity test) only among white men. Neither black nor oriental males exhibited this relationship whereas among the women (whether white or black) the fattest ones in the group were non-drinkers.
Roe's studies (2) have shown that the body weight of drinkers is strongly influenced by socio-economic status (which bears on diet) as well as how alcoholic excess occurs. A weight loss in intemperate individuals may have to do not only with the metabolic factors mentioned above in connection with repeated intakes of large quantities of alcohol but also with reduced intakes of solid food during periods of alcoholic excess. Another factor is the repercussions of alcoholic intoxication on intestinal absorption of nutrients. Unquestionably, there is a syndrome of intestinal malabsorption in the chronic alcoholic. The mechanism is complex. According to Jian and Modigliani (28) its dominant features are functional anomalies in exocrine pancreas and in intestinal mucous membrane which alcohol and malnutrition would act synergistically to produce.
II - ALCOHOL CONSUMPTION LEVEL NOT TO BE EXCEEDED
(G. PEQUIGNOT)*
The fact that the consumption of alcohol is legal in many countries rests on the assumption that there is some dose which has no effect. There is no scientific foundation for this. So we have a problem with no satisfactory solution: to define a level of alcoholic consumption not to be exceeded. The proposed level is usually arbitrary (29) and refers to only one of alcohol's effects. We might try to find part of the answer for three of alcohol's effects. Alcohol is at the same time:
* Research Director, INSERM, Paris
1. Problems with alcohol as a source of energy.
According to the above literature, it seems that the effect of very high doses (unquestionably toxic) on corpulence is nil or negative (4) (5) (6) (7) (8). However, the positive effect on corpulence of the addition of about 60 g of alcohol to the daily diet has been experimentally proven (3). So rations of up to 60 g of alcohol very probably contribute to the energy supply. These are the most frequently observed levels of consumption, even in countries where average consumption is the highest (30). So most drinkers should consider alcohol as a source of energy, the effect of which on body weight is all the more likely as a positive relationship between alcohol consumption and an alcohol-free diet in a sample of a French population (31). Therefore, where corpulence is excessive, the consumption of alcohol should be restricted, just as for sugar and fats. Where there is no excess weight and alcohol is added to an adequate diet, a weight gain may result. To avoid this consequence alcohol would have to be substituted for part of the energy intake, taking however into account the fact that alcoholic drinks do not supply any significant amounts of proteins, vitamins and mineral salts. In the most developed countries, which are the greatest consumers of alcohol, sedentary living brings the energy requirement down to about 2 000 to 2 500 calories in many adults. Under such conditions energy from saccharose should not exceed 10 percent or 200–250 calories (50–60 g). Isocaloric substitution of alcohol for half of all the maximum recommended saccharose intake would be 15–35 g of alcohol (one to three glasses a day). Explanatory studies suggest a maximum of 40 g of alcohol for pregnant women to prevent the risk of stillbirth (32). Lastly, alcohol-dependant hyperlipidemia calls for complete abstinence (33).
2. Problems of acute effects of alcohol.
The effect of alcohol on the central nervous system first shows up as initially low blood alcohol levels of .2–.3 g/l corresponding to ingestion of .3–.4 g/kg (about 20 g for a 70 kg individual), as the old studies had already shown (especially the simulated behaviour studies) (34) (35).
So apparently nil alcohol blood levels should be the goal where there is some risk of accident (on the job or on the road). For this reason it is advisable to avoid any activity involving the risk of accident for one and half hours after the absorption of a glass containing 12 g of alcohol, since alcohol is eliminated at about 0.100 g/kg/h (15). In many cases, any drinking before returning home from work should be avoided for safety's sake.
3. The problem of long-term toxic effects of alcohol.
In an interesting review of the question, Rydberg and Skerfving (1977) (36) suggest alcohol be considered a natural food contaminant and the regulations of the various FAO-WHO expert committees on acceptable levels of additives and contaminants applied.
Assessment of acceptable daily doses is based on an estimated maximum dose sustained throughout life without harmful effect. The definition of acceptable daily doses is obtained by applying a safety factor of 100 based on experiments with animals and 10 based on available data on man. According to the literature, the authors note that long-term toxic effect on the human liver is observed at rates of .5-3 g/kg and is probably certain at or above 1 g/kg, that is to say, 70 grams for an adult weighing 70 kg. Applying the factor of 10, an acceptable dose would be 7 g per day.
According to the most recent epidemiological data, any dose much beyond this figure appears hardly acceptable. Lelbach (37), based his findings on work covering systematic liver biopsies done on alcohol-dependent individuals in a deintoxication centre, suggested an upper limit of 35 g. The Ille and Vilaine study (30) (31) (38) on the risks of cirrhosis, cancer of the esophagus and delirium tremens from heavy alcohol consumption, did not pinpoint a clear threshold. Least risk in the three illness was noted for under 20 g of alcohol (the lowest threshold tested). Risk increases exponentially as declared doses increase. Other studies dealing with the risk of cirrhosis and calcifying pancreatitis (39) (40) also failed to reveal a clear toxicity threshold.
In all these epidemiological studies, risks are evaluated in accordance with statements made by the test subjects. It is quite likely that a certain underestimation of actual consumption entails a certain overestimation of risk as far as doses are concerned. But from a practical viewpoint it is both chimerical and futile to attempt to find the “real” consumption underlying the “declared” consumption as the latter is the only measurable factor. Since risk has been estimated in accordance with declared consumption, prevention can be exercised in accordance with declared consumption. Once we know that individuals with a declared consumption of 20–40 g of alcohol run much greater risks than those declaring 0–20 g, we want to try to get anyone declaring more than 20 g to cut down on his intake. A drink contains 8–15 g of alcohol (averaging about 12 g). So one should try to avoid drinking more than one and one-half glasses per day (not drinking every day since other days may well exceed this limit).
In conclusion the different aspects of the effects of alcohol covered here (balanced diet and corpulence, accident prevention, long-term prevention of toxicity) do not reveal a clear toxicity threshold.
BIBLIOGRAPHIE
1. Besoins énergétiques et besoins en protéines. 1973 Rapport d'un comité spécial mixte FAO/OMS d'experts. Org.mond.Santé.Sér.Rapp. techn., No. 522.
2. Roe, D.A. 1979 Alcohol and the diet, 229 pp., AVI Pub. Co., Westport, USA.
3. Belfrage, P., Berg, B., Cronholm, T., Elmquist, D., Hägerstrand, I., Johansson B., Nilsson-Ehle, P., Nordén, G., Sjövall, J. et Wiebe, T. 1973 Prolonged administration of ethanol to young, healthy volunteers: effects on biochemical, morphological and neurophysiological parameters. Acta Med. Scand., Suppl. 552, 5–44.
4. Pirola, R.C. et Lieber, C.S. 1972 The energy cost of the metabolism of drugs, including ethanol. Pharmacology, 7, 185–196.
5. Pirola, R.C. et Lieber, C.S. 1976 Hypothesis: energy wastage in alcoholism and drug abuse: possible role of hepatic microsomal enzymes. Am.J.Clin. Nutr., 29, 90–93.
6. Lieber, C.S., Jones, D.P. et DeCarli, 1965 M. Effects of prolonged ethanol intake: production of fatty liver despite adequate diets. J.Clin.Invest., 44, 1009–1021.
7. Pirola, R.C. et Lieber, C.S. 1975 Energy wastage in rats given drugs that induce microsomal enzymes. J.Nutr., 105, 1544–1548.
8. Lieber, C.S. 1977 Metabolic Aspects of Alcoholism, 308 pp., MTP Press, Lancaster U.K.
9. Yuki, T., Thurman, R.G., Schwabe, U. et Scholz, R. 1980 Metabolic changes after prior treatment with ethanol, evidence against an involvement of the Na+ K+ - activated ATPase in the increase in ethanol metabolism. Biochem.J., 186, 997–1000.
10. Israel, Y., Kalant, H., Orrego, H., Khanna, J.M., Phillips, M.H. et Stewart, D.J. 1979 Hypermetabolic state, oxygen availability, and alcohol - induced liver damage. In “Biochemistry and Pharmacology of Ethanol” E. Majchrowicz and E.P.Noble, éds.) Vol. 1, pp. 443–444, Plenum Press, New York.
11. Teschke, R. 1980 Leberschäden durch Alkohol. 201 pp., G. Thieme, Stuttgart.
12. Thurman, R.G. et Brentzel, H.J. 1977 The role of alcohol dehydrogenase in microsomal ethanol oxidation and the adaptive increase in ethanol metabolism due to chronic treatment with ethanol. Alcoholism.Clin. Exp.Res., 1, 33–38.
13. Wendell, G.D. et Thurman, R.G. 1979 Effect of ethanol concentration on rates of ethanol elimination in normal rats in vivo. In “Currents in Alcoholism” (M. Galanter, éd.) Vol. 5, pp. 81–89, Grune and Stratton, New York.
14. Feinman, L., Baraona, E., Matsuzaki, S., Korsten, M. et Lieber, C.S. 1978 Concentration dependence of ethanol metabolism in vivo in rats and man. Alcoholism. Clin.Exp.Res., 2, 381–385.
15. Eriksson, C.J.P. et Peachey, J.E. 1980 Lack of difference in blood acetaldehyde of alcoholics and controls after ethanol ingestion. Pharmacol. Biochem. Behav., 13, Suppl. 1, 101–105.
16. Bode, J.C. 1978 The metabolism of alcohol: physiological and pathophysiological aspects. J.Roy.Coll.Phycns. 12, 122–135.
17. Lumeng, L., Bosron, W.F. et Li T.K. 1980 Rate-determining factors for ethanol metabolism in vivo during fasting. in “Alcohol and Aldehyde Metabolizing Systems” (R.G. Thurman, éd.) Vol. 4, pp. 489–496, Plenum Press, New York.
18. Scholz, R. et Nohl, H. 1976 Mechanism of the stimulatory effect of fructose on ethanol oxidation in perfused rat liver. Eur.J.Biochem., 63, 449–458.
19. Bode, J.C. 1980 Factors influencing alcohol metabolism in man. Colloques INSERM, 95, 65–92.
20. Clark, E.R., Hughes, I.E. et Letley, E. 1973 The effect of oral administration of various sugars on blood ethanol concentrations in man. J.Pharm. Pharmac., 25, 319–323.
21. Beaugé, F., Mangeney, M., Nordmann, J. et Nordmann, R. 1980 Comparative study of the effect of amino acids on ethanol oxidation in isolated hepatocytes from starved and fed rats. In “Alcohol and Aldegyde Metabolizing Systems” (R.G. Thurman, éd.), Vol.4, pp. 393–402.
22. Krebs, H.A. et Stubbs, M. 1975 Factors controlling the rate of alcohol disposal by the liver. In “Alcohol Intoxication and Withdrawal” (M.M.Gross, éd.) Vol.2, pp. 149–161, Plenum Press.
23. Pawan, G.L.S. 1968 Physical exercise and alcohol metabolism in man. Nature, 218, 966–967.
24. Sautier, C., Bricout, J., Doucet, C., Flament, C., Kosiet, J. et Letonturier, P. 1981 Effet d'un travail musculaire sur l'oxydation du13C éthanol chez l'homme. A paraître.
25. Spritz, N. 1979 Review of the evidence linking alcohol consumption with liver disease and atherosclerotic disease. Am.J.Clin.Nutr., 32, 2734– 2738.
26. Abdushelishvili, G. 1976 The peculiarities of obesity frequency in the Georgian Republic. Proc. 10th. Internat.Congr.Nutrit. 1975, Kyoto, Japan, Victory-sha Press.pp.263–264.
27. Klatsky, A.L., Friedman, G.D., Siegelaub, A.B. et Gerard, M.J. 1977 Alcohol consumption among white, black or oriental men and women: Kaiser-Permanente multiphasic health examination data. Amer.J. Epidemiol, 105, 311–323.
28. Jian, R. et Modigliani, R. 1980 Absorption intestinale et alcoolisme chez l'homme. Gastroentérol.Clin.Biol., 4, 577–587.
29. Morgan, M.Y. 1979 Alcohol and the liver, Journ. of Hum. Nutr., 33, 350–356.
30. Pequignot, G., Tuyns, A.J. and Berta, J.L. 1978 Ascitic cirrhosis in relation to alcohol consumption, Inter.Journ. of Epidem., 7, 113–120.
31. Pequignot, G., and Tuyns, A.J. 1980 Compared Toxicity of Ethanol on Various Organs, in: Les colloques de l'I.N.S.E.R.M., Alcohol and gastrointestinal tract, 95, 17–32.
32. Kaminski, M., Franc, M., Lebouvier, M., du Mazaubrun, C., Rumeau-Rouquette, C. 1981 Moderate alcohol use and outcome of pregnancy, in: Pharmacology, Biochemistry and Behavior, supplement on Fetal Alcohol Syndrome ed. A.P. Streisguth and al. (sous presse).
33. De Gennes, J.L., Thomopoulous, P., Truffert, J., Labrousse, B. 1972 Hyperlipidémies dépendantes de l'alcool, Nutr. Metabol., 14, 141–158.
34. Goldberg, L. 1943 Quantitative studies on alcohol tolerance in man, Acta physiol. Scand. 5, Suppl. 16, 1–128.
35. Drew, G.C., Colquhoun, W.P., Long, H.A. 1958 Effect of small doses of alcohol on a skill resembling driving. Br.Med.J., 2, 993–998.
36. Rydberg, U. and Skerfving, S. 1977 The toxicity of ethanol. A tentative risk evaluation in: Alcohol intoxication and Withdrawal, Vol 3B, 403–419, ed. Gross M.M. Plenum publishing corporation.
37. Lelbach, W.K. 1975 Cirrhosis in the Alcoholic and its relation to the volume of alcohol abuse in: Medical Consequences of Alcoholism, ed. Seixas F.A., Williams, K. and Eggleston, S. Ann. N.Y.Acad.Sci. 252, 85–105.
38. Tuyns, A.J., Pequignot, G., Jensen, O.M. 1977 Le cancer de l'oesophage en IIIe et Vilaine en fonction des niveaux de consommation d'alcool et de tabac, Bulletin du cancer, 64, 45–60.
39. Durbec, J.P., Bidart, J.M., Sarles, H. 1980 Interaction between Alcohol and other Foodstuff. Epidemiological Aspects in: Les colloques de l'I.N.S.E.R.M., Alcohol and gastrointestinal tract, 95, 33–52.
40. Infante, F., Voirol, M., Raymond, L., Hollenweger, V., Conti, M.C., Loizeau, E. 1980 Alcohol, Tobacco and Nutriments consumption in liver cancer and cirrhosis patients in Geneva. In: Les colloques de l'I.N.S.E.R.M., Alcohol and gastrointestinal tract, 95, 53–58.
