The aim of this study was to assess a method able to analyze serum levels of risperidone (RIS) and its metabolite, 9-hydroxyrisperidone (9-OH-RIS), and to investigate possible relationships between changes in serum concentrations of these drugs and clinical measures, so to identify early markers of treatment response. The authors developed a sensitive and specific liquid chromatography-tandem mass spectrometry method to measure RIS and its metabolite in serum. Fifteen RIS-naive patients were admitted to an acute psychiatric care unit and treated with 4-6 mg/d oral RIS. At days 7 and 21 of hospital stay, serum levels were measured; clinical scales and serum prolactin were assessed. RIS and its metabolite were analyzed by a Q-Trap 2000 triple quadrupole/ion trap mass spectrometer in the multiple reaction-monitoring mode. Chromatographic separation was accomplished using a cyano column with an analytical run of 9 minutes. The calibration curve exhibited consistent linearity and reproducibility in the range 0-100 ng/mL for both analytes. Lower limit of quantification was 0.2 ng/mL; limit of detection, for a signal to noise ratio of 3, was 0.05 ng/mL for both analytes. Serum RIS and 9-OH-RIS levels increased at day 7, reaching a steady state, and remaining constant up to day 21. Scores on psychopathology rating scales decreased; serum prolactin and neurological rating scale for extrapyramidal effects rose at day 7 and remained stable thereafter. No correlation was found between serum concentration values, including sum and ratio of RIS and 9-OH-RIS, and any of the other clinical values (serum prolactin and clinical scales). These data indicate that clinical changes are related to the achievement of steady state levels of RIS and its metabolite and are maintained, but not continued, with continued RIS treatment. Therapeutic drug monitoring of RIS and its metabolites is not recommended as a routine procedure in patients with psychotic disorders.