Intravenous lipid emulsion has been suggested as treatment for local anaesthetic toxicity, but the exact mechanism of action is still uncertain. Controlled studies on the effect of lipid emulsion on toxic doses of local anaesthetics have not been performed in man. In randomized, subject-blinded and two-phase cross-over fashion, eight healthy volunteers were given a 1.5 ml/kg bolus of 20% Intralipid(®) (200 mg/ml) or Ringer's acetate solution intravenously, followed by a rapid injection of lidocaine 1.0 mg/kg. Then, the same solution as in the bolus was infused at a rate of 0.25 ml/kg/min. for 30 min. Electroencephalography (EEG) was recorded, and 5 min. after lidocaine injection, the volunteers were asked to report subjective symptoms. Total and un-entrapped lidocaine plasma concentrations were measured from venous blood samples. EEG band power changes (delta, alpha and beta) after the lidocaine bolus were similar during lipid and during Ringer infusion. There were no differences between infusions in the subjective symptoms of central nervous system toxicity. Lidocaine was only minimally entrapped in the plasma by lipid emulsion, but the mean un-entrapped lidocaine area under concentration-time curve from 0 to 30 min. was clearly smaller during lipid than Ringer infusion (16.4 versus 21.3 mg × min/l, p = 0.044). Intravenous lipid emulsion did not influence subjective toxicity symptoms nor affect the EEG changes caused by lidocaine.