INTERACTION BETWEEN DIETARY PATTERN AND ALCOHOL INTAKE ON THE RISK OFLIVER-CIRRHOSIS

In order to assess the inter-relationship between nutritional intake a nd alcohol consumption on the risk of liver cirrhosis we performed a h ospital-based retrospective case-control study. We enrolled 115 cases admitted to hospital for liver decompensation at their first diagnosis of liver cirrhosis and 167 hospital controls without evidence of live r disease admitted for acute diseases unrelated to alcohol intake. Dai ly alcohol intake and average nutrient intake were measured throughout the patient's life, using a reproducible questionnaire. No dose-effec t relationship was found between nutrient intake and risk of cirrhosis using classical association statistical methods. We then corrected th e intake of each nutrient for the total caloric intake and this energy -adjusted nutrient intake was used in a logistic regression model toge ther with alcohol intake, viral B and C hepatitis markers, age and gen der. Using this approach, carbohydrates intake were shown to have a pr otective effect on the risk of cirrhosis, whereas saturated lipid inta ke had a significant multiplicative effect on the risk associated with alcohol consumption. By comparison with the teetotalers category who had an average daily intake of saturated fatty acids lower than 40.3 g (reference category; OR=1), drinkers of more than 100 g ethanol per d ay showed ORs ranging from 14.2 (95% confidence interval 2.0-101.0) fo r consumers of less than 40.3 g fatty acid per day, to 39.0 (95% confi dence interval 5.0-305.1) for consumers of more than 40.4 g fatty acid per day. In conclusion we give additional evidence on the relationshi p between diet and risk of cirrhosis, whereby saturated lipid intake m ultiplies the risk associated with alcohol intake. However, caution sh ould be used to interpret such results, since they seem to suggest tha t diet but not a particular nutrient can modify the effect of alcohol on the risk of cirrhosis. The present lack of agreement on the mechani sms and the nutrients involved in the pathogenesis of alcoholic liver injury should stimulate wider epidemiological studies using modern nut ritional techniques.