It has been demonstrated that achieving therapeutic levels of cyclosporine (CsA) exposure in the first days posttransplant is critical for effective prevention of rejection. In patients receiving oral CsA, it has been shown that C(2)-monitoring is superior to trough (trough level [C(0)]) measurement. Intravenous administration may overcome the problem of CsA absorption dysfunction seen in some patients. Currently, little evidence is available concerning CsA exposure after intravenous application. Twenty de novo liver transplant recipients were given twice-daily 4-hour infusions of intravenous CsA, with full pharmacokinetic profiles undertaken during the first postoperative week. The greatest CsA exposure occurred during the period 2 to 4 hours after the start of infusion. The correlation between C(0) and area under the curve (AUC(0-12)) was r(2) = 0.18; the correlation between C(2) and AUC(0-12) was r(2) = 0.82. The best 2-point predictive model included both C(2) and C(4) (r(2) = 0.90). There was a poor correlation between CsA dose per kilogram of body weight and AUC(0-12) (r(2) = 0.19); total CsA dose also showed a weak relationship to exposure (r(2) = 0.37). When patients were divided according to initial or delayed graft function, there was good correlation between total CsA dose and AUC(0-12) (initial function, r(2) = 0.71; delayed function, r(2) = 0.86). In conclusion, previous discouraging results with intravenous CsA in liver transplant patients may have been due to a limited understanding of CsA pharmacokinetics. Our results show that C(2)-monitoring during 4 hour infusion provides a reliable indication of CsA exposure. Calculation of starting dose based on initial graft function is more precise than use of body weight. Using C(2)-monitoring to individualize dosing and function-based calculations of starting dose could be expected to improve clinical outcomes in patients receiving intravenous CsA.