Elsevier

Annals of Epidemiology

Volume 11, Issue 7, October 2001, Pages 458-465
Annals of Epidemiology

Original report
Coffee, Caffeine, and the Risk of Liver Cirrhosis

https://doi.org/10.1016/S1047-2797(01)00223-XGet rights and content

Abstract

PURPOSE: To evaluate the effect of the consumption of caffeine-containing beverages on the risk of symptomatic liver cirrhosis (LC).

METHODS: From 1994 to 1998, all the consecutive cirrhotic inpatients admitted in 19 collaborative hospitals for signs of liver decompensation in whom the diagnosis of liver cirrhosis was made for the first time (274 cases) and one or two gender, age, and place of residence pair matched individuals (458 controls) were recruited. Data on years of education, lifetime cigarette use, lifetime intake of alcohol- and caffeine-containing beverages, usual consumption of 180 food items, and on markers of hepatitis B and C viral infection were collected.

RESULTS: A statistically significant trend toward lowered cirrhosis risk with increasing exposure to coffee was observed. The LC odds ratios decreased from 1.0 (reference category: lifetime abstainers from coffee) to 0.47 (95% confidence interval: 0.20, 1.10), 0.23 (0.10, 0.53), 0.21 (0.06, 0.74), and 0.16 (0.05, 0.50) in 1, 2, 3, and 4 or more cups of coffee drinkers, respectively. There was no convincing evidence that coffee consumption modifies the effects of the known risk factors of liver cirrhosis (alcohol intake and viruses infection).

CONCLUSIONS: These findings support the hypothesis that coffee, but not other beverages containing caffeine, may inhibit the onset of alcoholic and nonalcoholic liver cirrhosis.

Introduction

The use of coffee has long been accompanied by debate about its possible risks and beneficial effects. Suspected adverse effects of coffee on the development of several pathological conditions (coronary heart disease, certain cancers, and reproductive abnormalities) remain controversial (1). On the other hand, a protective effect of coffee on liver cell damage caused by alcohol has been suggested by several studies. Inverse correlations between coffee consumption and serum liver enzyme levels have been reported in several studies 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12. Moreover, an association between coffee consumption and reduced risk of liver cirrhosis (LC) has been observed in a prospective epidemiological investigation (13), and in a case-control study (14).

Several questions still remain unresolved. First, coffee consumption might be a marker of other lifestyle factors that could act as protective factors. For example, several epidemiological studies suggests that certain dietary patterns might be associated with the onset of LC 15, 16, 17, and it is likely that there are important correlations among dietary and lifestyle factors (specific aliments, coffee, alcohol, smoking, etc.). Second, it is not yet clear whether the suspected protective effect of coffee is because of caffeine, the most prominent pharmacologically active substance contained in coffee (1), or to other substances. Third, to our knowledge, it has not been determined whether coffee acts only to inhibit the harmful effects of alcohol on the liver cell, or if similar effects are also present for other known determinants of LC (e.g., chronic infection by hepatitis B and C viruses–HBV and HCV, respectively).

We attempted to address these problems, using the data of the SIDECIR (Italian Study on Liver Cirrhosis Determinants) project, a new Italian collaborative case-control study (18). The main aims were: i) to estimate the independent effect of coffee, and of other beverages containing caffeine, on the risk of LC with adjustment for the known LC risk factors and for several lifestyle factors; ii) to explore the joint action of coffee, alcohol, HBV, and HCV infections, on the risk of LC.

Section snippets

Selection of Cases and Controls

From 1994 through 1998, 338 LC cases were recruited from those admitted to the Medical Departments of 19 Italian collaborative centers. Cases were admitted for signs of liver decompensation (ascites, jaundice, edema) or bleeding from ruptured esophageal varices, and in whom the diagnosis of LC was made for the first time. Forty-eight patients were excluded: 17 because of invalidating hepatic encephalopathy, 19 because of coexisting hepatocellular carcinoma, and 12 with a final ascertained

Results

High degrees of concordance were obtained between interviewers and supervisor (ranging from 0.79 to 0.85 for alcohol and caloric intakes, respectively), as well as between collaterals and patients (ranging from 0.78 to 0.89 for smoking and carbohydrates, respectively). This was also the situation within patients interviewed on two occasions (ranging from 0.87 to 0.93 for alcohol and caloric intakes, respectively) or interviewed with reference to two life periods (ranging from 0.75 to 0.92 for

Discussion

The well-established associations between alcohol intake and chronic HBV or HCV infections on the risk of LC are robustly confirmed by the present study. Further evidence that coffee drinking is associated with decreased LC risk is also provided. New findings include evidence that the observed inverse coffee/cirrhosis relationship: i) is not caused by the confounding effect of education, smoking habits, and dietary patterns, which were controlled in the analyses; ii) is probably not

Selected Abbreviations and Acronyms

ANOVA = analysis of variance

CI = confidence intervals

ELISA = enzyme linked immunosorbent assay

HBsAg = Hepatitis B virus surface antigen

HBV = Hepatitis B virus

HCC = Hepato cellular carcinoma

HCV = Hepatitis C virus

LC = liver cirrhosis

LRT = likelihood ratio test

OR = odds ratio

PBC = primary biliary cirrhosis

RIBA = recombinant immunoblot assay

SIDECIR = Italian Study on Liver Cirrhosis Determinants

Acknowledgements

The authors gratefully acknowledge the helpful suggestions of Dr. Astrid Nehlig. Supports for this study came from Institute for the Scientific Information on Coffee (ISIC, Paris) and from the Italian National Research Council (CNR No. 98.03003.CT04).

References (39)

  • O Nilssen et al.

    The Tromso Heart StudyDistribution and population determinants of gamma-glutamyltransferase

    Am J Epidemiol.

    (1990)
  • E Casiglia et al.

    Unexpected effects of coffee consumption on liver enzymes

    Eur J Epidemiol.

    (1993)
  • S Kono et al.

    Coffee and serum gamma-glutamyltransferaseA study of self-defense officials in Japan

    Am J Epidemiol.

    (1994)
  • O Nilssen et al.

    Seven-year longitudinal population study of change in gamma-glutamyltransferaseThe Tromso Study

    Am J Epidemiol.

    (1994)
  • F Pintus et al.

    Distribution and population determinants of gamma-glutamyltransferase in a random sample of Sardinian inhabitants

    Eur J Epidemiol.

    (1996)
  • K Poikolainen et al.

    Determinants of gamma-glutamyltransferasePositive interaction with alcohol and body mass index, negative association with coffee

    Am J Epidemiol.

    (1997)
  • K Tanaka et al.

    Coffee consumption and decreased serum gamma-glutamyltransferase and aminotransferase activities among male alcohol drinkers

    Int J Epidemiol.

    (1998)
  • A.I Klatsky et al.

    Alcohol, smoking, coffee, and cirrhosis

    Am J Epidemiol.

    (1992)
  • G Corrao et al.

    The effect of drinking coffee and smoking cigarettes on the risk of cirrhosis associated with alcohol consumptionA case-control study

    Eur J Epidemiol.

    (1994)
  • Cited by (0)

    Collaborative SIDECIR (Italian Study on Liver Cirrhosis Determinants) group: Morelli D., Moscatello M.R., Stefanini G.F. (Bologna); Chiesa R., Donato F., Tomasoni V. (Brescia); Guglielmini V., Manghisi O.G., Petruzzi J. (Castellana Grotte, Bari); Boncinelli L. (Como); Bocchia M., Ribotto P., Canevelli E. (Garbagnate, Milano); Isa L., Moriglioni M. (Gorgonzola, Milano); Sardi G.F., Monica M. (Lanzo Torinese, Torino); Fasoli, Baldacci (Legnano, Milano); Andri G., Migliosi G. (Manerbio-Leno, Brescia); Ajello A., Freni M.A., Spadaro A. (Messina); Gravina M., Longo G., Mangano C. (Messina); Silvani A., Bottelli R. (Milano); Fiorelli G., Fargion S., De Feo T. (Milano); Ascione A., De Luca M., Galeota Lanza A. (Napoli); Del Vecchio Blanco G., Federico A., Loguercio C. (Napoli); Burra P., Mioni D., Naccarato R. (Padova); Guarnone F., Pallavicini C., Vittadini G. (Pavia); Aricò S. (Torino); Salvagnini M. (Vicenza), Italy

    View full text