To evaluate the effect of donor age on graft and patient outcome after liver transplantation an analysis of a large-scale cohort study was performed at three tertiary referral liver transplant centers. Between April 1990 and June 1994, 772 adults underwent an initial single-organ liver transplantation. The age of the donors averaged 35 years;193 (25%) were 50 or above, the age used to define "older" donors. Groups were compared for demographic, clinical, and biochemical features. Outcome was measured using results of biochemical tests after transplantation and by graft and patient survival. Compared with younger donors, older donors were more commonly women (59% vs. 33%:P < or = .001) and died of central nervous system causes (79% vs. 28%) as opposed to trauma (13% vs. 63%:P < or = .001). The recipients of the two groups of donor livers did not differ in important respects. However, intraoperatively, livers from older donors were more likely to be assessed as either "poor" or "fair" as opposed to "good" (17% vs. 4%:P < or = .001) by the harvesting surgeon and to have initial "poor" or "fair" bile production (29% vs. 18%:P < or = .001). During the first week postoperatively, the serum aminotransferase and bilirubin levels and prothrombin times were higher in recipients of older than those of younger donor livers. During follow-up, graft survival was less for recipients of older donor livers at 3 months (81% vs.91%:P = .0001) and at 1 (76% vs. 85%:P = .007) and 2 years (71% vs. 80%:P = .005); patient survival showed similar though less marked differences. This association of donor age and poorer graft survival persisted after adjusting for many variables using bivariate and multivariate analyses. Importantly, however, the association with poor graft survival was largely among recipients of older donor livers, the quality of which was assessed as fair or poor by the harvesting surgeon; recipients of older donor livers assessed as good had a retransplant-free survival similar to that of younger donor livers (87% vs. 91% at 3 months). Thus, use of older donor livers, the quality of which are judged to be good by the harvesting surgeon, is not associated with a decrease in patient or graft survival after liver transplantation. differences. This association of donor age and poorer graft survival persisted after adjusting for many variables using bivariate and multivariate analyses. Importantly, however, the association with poorer graft survival was largely among recipients of older donor livers, the quality of which was assessed as fair or poor by the harvesting surgeon; recipients of older donor livers, the quality of which was assessed as fair or poor by the harvesting surgeon; recipients of older donor livers assessed as good had a retransplant-free survival similar to that of younger donor livers (87% vs. 91% at 3 months). Thus, use of older donor livers, the quality of which are judged to be good by the harvesting surgeon, is not associated with a decrease in patient or graft survival after liver transplantation.