The pharmacokinetics of remoxipride, a new selective dopamine-D2 receptor antagonist with antipsychotic action, was evaluated in eight elderly psychiatric patients with tardive dyskinesia (TD). The daily oral doses of remoxipride were gradually increased from 50 mg per day to 200 mg t.i.d. over 2 weeks. The pharmacokinetics following the initial 50 mg dose (day 1) and the last 200 mg dose (day 15) of the drug were compared in serial samples. Plasma prolactin concentrations were assessed at the same time points. The area under the total plasma concentration versus time curves (AUC) of remoxipride increased proportionally with dose from day 1 to 15. The mean "dose corrected" AUC values for the total concentrations were 96.8 at day 1 (4×24.2, 50 mg single oral dose) and 92.2 μmol·h/l at day 15 (200 mg). The unbound fraction of remoxipride calculated on AUC was slightly higher on day 15 (20%) than on day 1 (15%) (P<0.05), indicating slightly concentration-dependent protein binding of the drug. The mean elimination half-life of total remoxipride was slightly longer on day 15 than day 1 (7.5 versus 5.3 h, P<0.01) The corresponding half-lives for the unbound concentrations were 6.4 and 3.9 h, respectively (P<0.01). The pharmacokinetics of remoxipride is similar in these TD patients and in non-TD patients in previous studies. Following repeated administration of remoxipride, tolerance to the prolactin-releasing action of remoxipride is observed. In plots of the relationship between the concentration of remoxipride and the prolactin increase during a dosage interval, the curve on day 15 was positioned markedly to the right of that on day 1. This indicates a decreased sensitivity to the prolactin releasing action of remoxipride following continuous treatment. © 1991 Springer-Verlag.