Mortality for liver disease in patients with HIV infection: a cohort study.

We undertook this study to assess the association between the various potential causes of liver disease in HIV-seropositive patients and mortality due to liver failure. Three hundred and eight in-hospital deaths were observed from 1987 to December 1995 in a prospectively followed cohort of 1894 HIV-seropositive patients. For each study subject, clinical data were evaluated to assess whether liver failure had substantially contributed to mortality. A case control study nested in the cohort was then performed, which compared demographic and clinical variables observed at enrollment and during follow-up between patients who died for liver disease as the main or concurrent cause of death (cases) and those who died as a result of illness related to AIDS or other causes (controls). Among 308 in-hospital deaths, liver failure was found the cause of death in 35 patients (12%); in 16 cases, it was the primary cause and in 19 cases it was concurrent. Multivariate analysis showed that in-hospital liver-disease-related mortality was independently associated with hepatitis B surface antigen reactivity (odds ratio [OR], 9; 95% confidence interval [CI], 3.8-21.7) and history of alcohol abuse (OR, 2.3; 95% CI, 1-5.2). Prevention and treatment of hepatitis B virus infection and alcohol intake are management priorities in HIV-seropositive patients.