It has been proposed that evaluation of alcoholic patients for liver transplantation should comprise a multidisciplinary program including medical, surgical, and psychiatric assessment. This article presents the outcome in 99 alcoholic persons evaluated from 1985 through 1989. Alcoholism was defined as alcohol dependence or abuse according to the DSM-III-R. Forty-five patients (43%) were considered suitable candidates and underwent transplantation. Actuarial survival in this group was 78% and 73% at 12 and 24 months, respectively, and did not differ from that observed in nonalcoholic recipients. Return to alcohol use after transplantation has been documented in 5 persons, 2 of whom have had episodes of uncontrolled drinking. Of 54 patients not selected for transplantation, 17 were considered too well and suitable for alternative therapy. Actuarial survival was 93% at 12 and 18 months but declined to 59% at 24 months. Nineteen patients were medically unsuitable and had very poor survival; survival was 35% at 3 months and 0% at 12 months. Seventeen patients were not selected on psychiatric grounds. Actuarial survival in this group (65% at 12 months, 43% at 18 months) was significantly less than that in alcoholics with transplants (P less than 0.05). There were no differences in age, sex ratio, degree of hepatic dysfunction, or duration of abstinence between alcoholics with transplants and those considered psychiatrically unsuitable. It was concluded that liver transplantation is successful in selected patients with alcoholic cirrhosis, that patients who are not selected because they are "too well" need continuing surveillance as a significant decline in survival can occur 18-24 months after initial evaluation and, finally, that patients not selected on psychiatric grounds have a significantly worse survival rate than those with transplants.