Cyclosporine A (CsA) is widely used after organ transplantation. Its narrow therapeutic window and large pharmacokinetic variability makes therapeutic drug monitoring (TDM) demanding and frequent dose adjustments are needed, especially early after transplantation. The aim of the present pilot study was to compare accuracy of CsA TDM by experienced clinicians against a computer-assisted dosing model. Renal transplant recipients on CsA, prednisolone, and mycophenolate were included 2 weeks after transplantation, randomized (1:1) to either computer dosing (MAP-BE) or control (CONTR) and followed for at least 8 weeks. A maximum a posteriori probability Bayesian estimation method, applying a population pharmacokinetic model and the POSTHOC option in nonlinear mixed effects modeling, was used to individualize CsA doses in the MAP-BE group. Forty patients (31 men, 27.5% living donor) between 28 and 80 years were included. A total of 798 CsA concentration measurements and adherent dosing evaluations/adjustments were performed. During the entire study, blood concentrations were on average 10% +/- 5% from the predefined therapeutic target range in the MAP-BE group, as compared with 13% +/- 8% in the CONTR group (P = 0.042). However, there was no significant difference between groups regarding the percentage of CsA concentrations truly within the therapeutic windows [MAP-BE: 37% +/- 17%, CONTR: 33% +/- 15% (P = 0.57)] or in CsA dose [MAP-BE: 3.55 +/- 0.8, CONTR: 3.90 +/- 0.9 mg/kg/d (P = 0.26)]. Acute rejections were present in 4 and 3 patients, respectively (P = 1.00). The computer-assisted TDM-targeted CsA blood concentrations significantly better than experienced transplant physicians. A possible favorable effect on short- and long-term outcome needs to be verified in further, properly powered, clinical trials.