A double-blind, randomized, placebo-controlled cross-over multi-center study was conducted to evaluate the efficacy and safety of gabapentin in the treatment of neuropathic pain caused by traumatic or postsurgical peripheral nerve injury, using doses up to 2400 mg/day. The study comprised a run-in period of two weeks, two treatment periods of five weeks separated by a three weeks' washout period. The primary efficacy variable was the change in the mean pain intensity score from baseline to the last week of treatment. Other variables included pain relief, health related quality of life (SF-36), interference of sleep by pain, Clinician and Patient Global Impression of Change, and adverse effects. Nine centers randomized a total of 120 patients, 22 of whom withdrew. There was no statistically significant difference between the treatments for the primary outcome efficacy variable. However, gabapentin provided significantly better pain relief (p=0.015) compared with placebo. More patients had at least a 30% pain reduction with gabapentin compared with placebo (p=0.040) and pain interfered significantly less with sleep during gabapentin treatment compared with placebo (p=0.0016). Both the Patient (p=0.023) and Clinician (p=0.037) Global Impression of Change indicated a better response with gabapentin compared with placebo. Gabapentin was well tolerated. The most common adverse effects were dizziness and tiredness.