Rabbits (n=23) received intravenous bolus medetomidine at 100 mug/kg. Prior to medetomidine administration, heart and respiratory rates were measured, arterial blood was collected and analysed for plasma cortisol, glucose and albumin concentrations. Fifteen minutes after medetomidine administration, heart and respiratory rates were measured again and sedation was scored. The rabbit was afterwards anaesthetized with 20 mg/kg ketamine administered intravenously to enable spinal tap and heart puncture. Cerebrospinal fluid (CSF) was collected (this occurred 20 min post medetomidine administration) and analysed for medetomidine concentration. Blood was collected by heart puncture immediately after the spinal tap and analysed for serum medetomidine concentration. Cerebrospinal fluid medetomidine concentration correlated negatively with sedation. Serum medetomidine correlated positively with CSF medetomidine concentration. Cerebro-spinal fluid medetomidine was 17 +/- 13% of serum medetomidine concentration. Plasma cortisol and glucose concentrations correlated negatively with serum medetomidine. We conclude that after an intravenous bolus administration of a low sedative dose of medetomidine to rabbits; CSF concentration of the drug correlate negatively with sedation and that this may be because of the fact that only the free and unbound medetomidine may be available for detection in the CSF, the concentration of medetomidine detected in the CSF was much lower than that in blood and a positive correlation exists between CSF and serum medetomidine concentrations. Stress may have some effect on the distribution or metabolism of medetomidine in rabbits.