We have previously reported that low hepatic tissue cyclosporine levels correlate with early cellular rejectionafter liver transplantation. The aim of this study is to determine whether there is a similar relationship in patients treated with FK506. Twenty-five liver biopsies were performed in 10 patients immunosuppressed with FK506 without cellular rejection: day 7 = 10; day 14 = 3; day 21 = 9; day 28 = 1; day 35 = 1; and day 42 = 1. These 10 patients without cellular rejection were compared with 7 patients immunosuppressed with FK506 with cellular rejection who underwent a total of 23 liver biopsies, including 9 biopsies that showed rejection: day 7 = 4; day 14 = 2; day 21 = 1; day 28 = 1; and day 49 = 1. There was no significant difference between the nonrejection and current rejection groups in the median plasma concentration of FK506, 0.9 ng/mL versus 0.9 ng/mL (P = 0.50). In contrast, the median hepatic tissue concentration of FK506 was significantly higher in the nonrejection group than it was in the current rejection group, 144 ng/g versus 48 ng/g (P = 0.02). In the current rejection group, 7 of 9 hepatic tissue concentrations of FK506 were <100 ng/g, and in the nonrejection group, 18 of 25 hepatic tissue concentrations were >100 ng/g. Low hepatic tissue concentrations of FK506 correlate with the occurrence of early cellular rejection after liver transplantation, in contrast to plasma concentrations of FK506. A hepatic tissue concentration of FK506 <100 ng/g is 78% sensitive and 72% specific for cellular rejection. These findings support the hypothesis that low hepatic tissue concentrations of immunosuppressives may lead to cellular rejection after liver transplantation. (Hepatology 1995;21:70–76).