Different methods of tourniquet release have been proposed to decrease the concentrations of local anaesthetic released into the systemic circulation at the end of intravenous regional anaesthesia. The effect of releasing the tourniquet intermittently with 5 seconds (group I) and 30 seconds (group II) deflation periods or at once (group III) was studied in 25 adult patients after intravenous regional anaesthesia with 40 ml of 0.5% lidocaine. The venous plasma lidocaine concentrations from the contralateral arm were measured by gas chromatography. There was no leakage of lidocaine from the occluded arm into the systemic circulation. The mean maximum plasma lidocaine concentration in group I 1.99 +/- 1.45 (SD) microgram/ml, in group II 1.33 +/- 0.54 microgram/ml and in group III 1.56 +/- 0.88 microgram/ml (P greater than 0.05) was below the toxic concentrations reported in the literature. There were subjective complaints such as dizziness and ringing in the ears in 4 out of the 7 patients in group I, in 2 out of the 9 patients in group II and in one of the 9 patients in group III (P greater than 0.05). There was no correlation between the duration of tourniquet time (range 12-87 minutes) and the maximum plasma lidocaine concentration. The intermittent release of the tourniquet did not decrease the venous plasma lidocaine concentrations in the contralateral arm; neither did comparing the lidocaine pharmacokinetics in 5 patients of group II after tourniquet release and in the 5 healthy volunteers after a single 100 mg intravenous lidocaine injection reveal any differences.