Independent risk factors for hepatocellular carcinoma in French drinkers

• Published in: Hepatology (Baltimore, Md.) Vol. 13; no. 5; p. 896
• Main Authors: Poynard, T, Aubert, A, Lazizi, Y, Bedossa, P, Hamelin, B, Terris, B, Naveau, S, Dubreuil, P, Pillot, J, Chaput, J C
• Format: Journal Article
• Language: English
• Published: United States 01.05.1991
• Subjects: Age Factors; Alcohol Drinking - adverse effects; Algorithms; Carcinoma, Hepatocellular - etiology; Female; France; Hepacivirus - immunology; Hepatitis Antibodies - blood; Hepatitis B Antibodies - blood; Hepatitis B Surface Antigens - blood; Humans; Liver Cirrhosis - complications; Liver Neoplasms - etiology; Male; Prospective Studies; Retrospective Studies; Risk Factors; Sex Factors; France
• ISSN: 0270-9139
• DOI: 10.1002/hep.1840130516

The aim of this study was to assess whether markers of hepatitis B virus or hepatitis C virus infection are independent risk factors for hepatocellular carcinoma in drinkers after adjustment for three known risk factors: cirrhosis, age and male sex. Among 2,015 consecutive drinkers admitted, hepatitis C virus antibodies were found by sensitive radioimmunoassay in 1,259. The following five factors have been identified and ranked as risk factors for hepatocellular carcinoma in unidimensional and regression analysis: cirrhosis (p less than 0.001), age (p less than 0.001), male sex (p less than 0.001), presence of HBsAg (p less than 0.001) and presence of hepatitis C virus antibodies (p less than 0.03). Among drinkers with cirrhosis, the patients with hepatocellular carcinoma were older (64 +/- 11 yr vs. 56 +/- 9 yr; p less than 0.001), were more often male (93% vs. 65%; p less than 0.0001) and had higher prevalence of HBsAg (9% vs. 2%; p = 0.02) and hepatitis C virus antibodies (41% vs. 26%; p = 0.02). A simple algorithm permitted us to identify a high-risk population of drinkers: the male cirrhotic patient older than 50 yr. The relative risk of hepatocellular carcinoma in this selected population was 17.7 (95% confidence interval = 9.0 to 37.5; p less than 0.0001). From a pragmatic point of view, the detection of HBsAg or hepatitis C virus antibodies, although independently associated with hepatocellular carcinoma, is not useful in increasing the diagnostic value of this algorithm because of the poor sensitivity of these tests. The weak relationships observed between hepatitis C virus antibodies and hepatocellular carcinoma needs confirmation by more accurate tests.