Aim: To evaluate 5-year survival predictive factors in hospitalised patients with excessive alcohol intake and cirrhosis, including in a multivariate analysis the severity of the liver disease, gastrointestinal bleeding, concomitant viral B or C infection, smoking status, presence of alcoholic hepatitis at inclusion and abstinence from alcohol during follow-up.
Methods: In a non-concurrent cohort study, 122 patients with excessive alcohol intake and cirrhosis were followed up at least five years or till death. Two patients were lost to follow-up.
Results: The 5-year survival rates were 43% in the 122 patients and 66%, 50% and 25% in Child-Pugh class A, B and C patients, respectively. In multivariate analysis, age (P = 0.01), Child-Pugh score (P = 0.0001), gastrointestinal bleeding (P = 0.01), presence of HBs Ag and/or anti-HCV (P = 0.03), smoking (P = 0.01), absence of histologically proven alcoholic hepatitis (P = 0.05) and persistent alcohol intake (P = 0.002) were associated with significantly increased risk ratios of death.
Conclusions: In hospitalised patients with excessive alcohol intake and cirrhosis: (1) age, liver failure, gastrointestinal bleeding, concomitant viral B or C infection and persistent alcohol intake are independent poor prognostic markers, (2) smoking may contribute to the aggravation of cirrhosis, and (3) alcoholic hepatitis, being a potentially reversible cause of liver failure, has a favourable prognostic significance.