The objective of this study was to determine the concentration-electroencephalogram (EEG) relationships for midazolam, a full-agonist benzodiazepine ligand, on multiple occasions in individual rats, and to examine the effect of chronic midazolam exposure on that relationship. Rats were chronically instrumented with venous and arterial cannulas, and cortical EEG electrodes. The rats received either: 7 days of midazolam 10 mg kg(-1) intravenously once a day (midazolam group); or midazolam on days 1 and 7 and vehicle on days 2-6 (vehicle group). Concentration-effect relationships were determined on days 1, 4 and 7 from multiple blood and EEG samples before and after the administration of the midazolam dose. The concentration-EEG effect relationships were consistent with a sigmoidal Emax (maximal effect) model. No differences in pharmacokinetic or pharmacodynamic parameters were found between day 1 and day 7 in either group. However, in the midazolam group, both the fraction unbound of midazolam in serum and the EC50 (concentration at half-maximal effect) for free midazolam increased from days 1-7 by 35 +/- 3% and 54 +/- 25%, respectively (means +/- s.d., P< 0.05). This may be related to decreased serum albumin levels in the midazolam group (-19+5%, P < 0.05) which, in turn, could be explained by the sedation associated with daily midazolam treatment. We concluded that concentration-EEG effect relationships can be studied on multiple occasions in individual animals, reducing animal use and variability. A modest degree of tolerance to midazolam was found with this paradigm, the effect only being evident after correction for the fraction unbound of midazolam.