Despite two decades of use, there are limited data on the best way to administer and monitor cyclosporine for orthotopic liver transplantation (OLT). The present study was undertaken: 1) to establish the safety of a new formulation of cyclosporine, Neoral, 2) to determine if treatment with Neoral will improve the results of liver transplantation and 3) to study the relationship between pharmacokinetic parameters and clinical outcomes after OLT. A double-blind, randomized, comparison of Sandimmune and Neoral was conducted at 5 Canadian centers in 188 consecutive adults undergoing OLT. Patients were induced with intravenous cyclosporine (CsA) then switched to Neoral or Sandimmune. Dose adjustments were made daily, or as needed, to reach a target trough CsA level (C0) of 350 ng/mL in both groups. Pharmacokinetic studies were performed on days 5, 10, 15 and 30 after transplantation. The Neoral group stopped intravenous CsA earlier (p < 0.0001), and these patients required a lower median daily oral dose (p < 0.01) to maintain comparable trough CsA levels. Five Sandimmune patients, but no Neoral patients discontinued the study because of the inability to reach target trough levels of CsA within the prescribed time (p < 0.05). At 4 months, there were no differences between the two groups with respect to patient survival, graft survival or rejection-free survival. The incidence of serious adverse events was also similar and did not correlate with CsA profiles. The Neoral group had a higher area under the drug concentration curve (AUC) and peak CsA levels (Cmax). There was a correlation between freedom from graft rejection and both AUC and Cmax at days 5 and 10 post-transplant. In contrast, there was a poor correlation between C0 and graft rejection. In summary, Neoral appears to be safe and well tolerated by patients. Cmax and/or AUC maybe better markers for monitoring cyclosporine based immunosuppression after liver transplantation.