Between 1988 and 1990, the frequency of liver transplantation in the United States increased by 57%. During this same period, the number of transplant centers performing this procedure increased from 58 to 80. Despite this increase, only 15 centers reported a total of at least 100 procedures during these 3 years, compared to 25 centers that performed 12 or less liver transplantations. Recipient characteristics have been changing over time: a larger proportion of recipients were males in 1990 than in 1988 or 1989. The distribution of recipients changed dramatically; the median age increased by 4 years, due to an increased proportion of transplantations among those age 40 and older and a decrease in children younger then age 10. Another major change was in functional status; in 1988 and 1989, over half of the recipients were hospitalized while awaiting transplantation, but this was reversed in 1990, when the majority of patients was at home awaiting transplantation. Furthermore, the proportion of patients in the highest functional class more than tripled. Alcoholic liver disease, which in 1989 became the most common primary liver disease of patients undergoing liver transplantation, continues to be the indication for an increasing number of recipients. The proportion of recipients with biliary atresia and primary biliary cirrhosis, the most common diagnoses in 1988, continues to decrease. Most of the mortality was noted in the first 6 months, when overall cumulative patient mortality was about 20%, half of which occurred in the first 4 weeks after OLTX. The cumulative 3-year posttransplant survival rate was 67%. Similarly, cumulative retransplant-free survival rates were 84% at 1 month and 58% at 3 years. As previously described (1), recipient factors associated with survival included age, UNOS description, diagnosis, and ABO matching. Older recipients, those with poorer functional status at time of transplantation, recipients with either fulminant liver failure or malignancies, and those who received a graft from an ABO-compatible or -incompatible donor, had the worst survival rates. Furthermore, in the current analysis we found that the outcome of recipients was affected adversely by grafts from female donors. Racial differences were noted, but the large quantity of missing data precluded definitive statements regarding any association with survival. Both recipient and donor ages were significant prognostic factors. For adults in the multivariate model, increasing recipient age was associated with higher mortality. Among children, however, younger donor age seemed to have an adverse effect on recipient survival. Donor characteristics also changed during this period.(ABSTRACT TRUNCATED AT 400 WORDS)