Systemic lidocaine has been reported to be effective in treating several neuropathic pain syndromes. Few reports relate plasma lidocaine concentration to analgesia and the available studies have been complicated by labile plasma lidocaine concentrations. We used a computer-controlled infusion pump (CCIP) to target and maintain stable plasma lidocaine concentrations and study the effect of intravenous lidocaine on (1) pain scores, (2) current perception thresholds, (3) side effects, and (4) pain distribution in patients suffering from peripheral nerve injury pain. This study used a randomized double-blind placebo-controlled design. Eleven patients suffering from neuropathic pain after peripheral nerve injury received both a lidocaine and saline infusion in separate study sessions. The order of the study sessions was randomized and separated from each other by 1 week. The CCIP was programmed to target plasma lidocaine concentrations of 0.5, 1, 1.5, 2, and 2.5 micrograms/ml, each held for 10 min. Pain scores and pain distribution were assessed in the painful area, and electrical current perception thresholds (CPT) of the ring finger were measured using a cutaneous perception threshold neurometer (Neurometer CPT, Neurotron, Baltimore, MD). Side effects were recorded at fixed intervals. Plasma lidocaine concentrations were measured at 4 and 9 min after each step increase in infusion and correlated with the observed effects. Saline infusion had no effect. However, with lidocaine there was a significant plasma concentration-dependent decrease in pain scores starting at 1.5 micrograms/ml. This effect typically corresponded with a decrease in the size of the receptive field to which the pain was referred. For the electrical stimulus, there was no significant effect on cutaneous perception at 2000-Hz stimulation at the highest concentration examined; however, there was a significant increase in thresholds at 250-Hz (starting at 1.5 micrograms/ml) and 5-Hz (starting at 1.0 micrograms/ml) stimulation. There were no serious side effects. In all, 54.5% of patients reported lightheadedness (average plasma lidocaine concentration: 1.5 micrograms/ml) and one patient reported nausea (2.3 micrograms/ml). The computer-controlled delivery of intravenous lidocaine results in relatively stable plasma concentrations which allows a more thorough evaluation of the relationship between plasma concentration and patient response. This administration methodology for intravenous lidocaine may prove to be a valuable clinical and research tool.