Semi-micro column high-performance liquid chromatographic method with ultraviolet detection for the determination of triazolam (TZ) in rat plasma and brain microdialysate is described. The separation was achieved on a 250x1.5 mm, i.d. C 1 8 column and the column effluent was monitored at 222 nm. The detection limits at a signal-to-noise ratio of 3 obtained using spiked plasma and artificial cerebrospinal fluid were 2.1 and 0.7 ng/ml, respectively. The method was applied to drug-drug interaction study of TZ with itraconazole (ITZ). The peak concentration (C m a x ) and the area under the curve (AUC) of TZ in brain microdialysate after simultaneous administration of TZ (2.5 mg/kg, intravenously (i.v.)) and ITZ (25 mg/kg, p.o.) to rats increased 3.4-folds (P<0.001) and 2.9-folds (P<0.001), respectively, compared to those of TZ alone. Also, the AUC of TZ in plasma increased 2.6-folds and remarkable delay in its elimination half-life (t 1 / 2 ) was observed. The concentrations of TZ in brain microdialysate and plasma were also measured after single administration of TZ (2.5 mg/kg, i.v.) to rats pretreated with daily administration of ITZ (25 mg/kg, p.o.) once a day for a week. There was no significant difference in TZ's C m a x in both ITZ treatments (P>0.2) however its t 1 / 2 after the daily pretreatment with ITZ was significantly increased (P<0.05). In plasma, the AUC of TZ after daily pretreatment of ITZ was lower than the single combined treatment, but significantly different from TZ's AUC in the absence of ITZ (P<0.05). As a result, single simultaneous administration of TZ with ITZ and single administration of TZ after daily pretreatment with ITZ to rats, ITZ seriously interfered with the pharmacokinetic parameters of TZ in plasma and brain micodialysate.