The purposes of this study were to determine among a cohort of long-term alcoholic survivors after liver transplantation (1) the incidence of alcohol use, (2) its effect on allograft integrity and extrahepatic health, and (3) the validity of the pretransplant alcohol prognosis screening process. Retrospective clinical cohort study of all alcoholic patients undergoing orthotopic liver transplantation at a single center from February 1987 until January 1991 with follow-up through December 1994, giving a median duration of follow-up of 63 months (range, 6-89 months). Multidisciplinary liver transplantation program at a tertiary-care academic medical center. Fifty alcoholic, long-term liver transplant recipients. The frequency of alcohol relapse, defined as any alcohol use in the period after transplantation, was determined by two questionnaire studies and by clinical follow-up. Allograft integrity was assessed by coded review of serial percutaneous allograft biopsies. Potential systemic effects of alcohol relapse were assessed by chart review. The alcohol prognosis screening process was evaluated by retrospectively comparing pretransplant estimates of putative indicators of alcoholism prognosis in posttransplant alcohol users and abstainers. Thirty-three recipients (66%) consistently denied any alcohol use throughout the duration of posttransplant follow-up, whereas 17 (34%) were identified as having consumed alcohol at least once since the transplant. There were no significant differences at the time of evaluation between abstainers and alcohol users in age, sex distribution, severity of liver dysfunction, median duration of abstinence, or University of Michigan alcoholism prognosis score. The median interval from transplantation to alcohol relapse was 17 months, with a range of 3 to 45 months. Recurrent alcohol use was associated with significant medical complications sufficient to require admission to the hospital in 6 patients. One patient died of graft dysfunction, noncompliance with immunosuppressant medications, and presumed graft rejection while drinking. Mild or progressive hepatitis, which was the most common abnormality in posttransplant liver biopsy findings, was equally distributed between both alcohol users and abstainers and sometimes occurred in the absence of antibody to hepatitis C virus antibodies. There was a similar frequency of biopsy-proven acute cellular rejection in alcohol users and abstainers. Typical histological features of alcoholic liver injury were present in posttransplant biopsies from 1 alcohol user only. Alcohol use by alcoholics is uncommon in the first 5 years after liver transplantation, and alcohol-associated liver injury is unusual. Mild nonspecific hepatitis is common in both alcohol users and nonusers alike. Among a small subset of alcoholic transplant recipients, drinking behavior after liver transplantation is associated with considerable morbidity, requiring hospital admissions and occasionally leading to graft loss and death.