In order to investigate the influence of general anaesthesia on the pharmacokinetics of bupivacaine 0.5% after a single, epidurally injected dose of 20ml (= 100mg), 5 patients, scheduled for surgical procedures appropriate for epidural anaesthesia without the need for combination with general anaesthesia, were studied. In all patients an arterial, central venous and intrathecal catheter were inserted for sampling of, respectively, arterial and central venous blood and cerebrospinal fluid (CSF), at regular intervals up to 12 hours after administration. Bupivacaine was analysed by gas chromatography. The drug appeared after 2 minutes in the plasma and CSF samples, and was still detectable after 12 hours. There were no significant differences between the bupivacaine concentrations in the arterial and central venous plasma samples at all times. The same applied for the calculated plasma pharmacokinetic parameters at all intervals. The bupivacaine concentrations in the CSF were many dozen times higher than in the plasma samples at all times. There was no clear evidence of a pulmonary retention of bupivacaine in this study. From a pharmacokinetic point of view the epidural anaesthesia was largely spinal anaesthesia.