Epidural clonidine increases acetylcholine (ACh) concentrations in cerebrospinal fluid (CSF) in humans, and experiments in animals support a cholinergic link in spinal alpha(2)-adrenoceptor-mediated antinociception. The purpose of the present study was to evaluate whether intravenous (IV) clonidine is also able to increase CSF ACh in humans. Accordingly, we studied 20 patients scheduled for resection of an acoustic neuroma under general anesthesia. Anesthesia was induced with propofol and maintained with propofol and N2O. After induction, an intrathecal catheter was inserted at the L3-4 interspace. Patients were then assigned, in a random, blind manner to receive either a bolus of 1 mu g/kg intrathecal (IT) clonidine and an IV infusion of saline (n = 10) or an IV infusion of 4 mu g/kg clonidine given in 20 min and an IT injection of saline (n = 10). CSF samples for ACh and clonidine concentration determination were drawn immediately before IT injection (time - 20), at the end of the IV injection (time 0), then every 10 min thereafter. CSF ACh concentrations were determined by high-pressure liquid chromatography and CSF clonidine by radioimmunoassay. There was no significant difference between the groups with respect to age, gender, weight, and ASA physical status. IT but not IV administration of clonidine increased the CSF ACh concentration. We conclude that IV administration of four times the dose of clonidine delivered spinally failed to induce a significant increase of ACh in the CSF. These observations indicate that the analgesic effects observed after IV clonidine administration are not mediated by a cholinergic mechanism at the spinal level.