The aim was to compare the analgesic effectiveness and adverse effect incidence of oral dextromethorphan (DM) with placebo in patients with neuropathic pain. The first 10-day treatment period was a multiple-dose double-blind randomised controlled cross-over comparison of 13.5 mg of DM 3 times a day (t.d.s.) with placebo t.d.s.: 5 treatment pairs, each pair 1 day DM and 1 day placebo. The second 10-day treatment period used 27 mg of DM t.d.s. vs. placebo, with the same design. The study incorporated a 5 pair n-of-1 design for each of the 2 doses of DM. Patients took the study medication in addition to any pre-existing analgesic regime. Patients who reported benefit could continue with DM after the study. Nineteen patients with chronic neuropathic pain were studied over two 10-day treatment periods. Outcome measures were pain intensity, pain relief, adverse effects, mood, sleep and global rating of treatment. These were recorded by daily patient diaries and by clinic assessments before and after each treatment period. There were no significant differences between DM and placebo on any of the clinic assessment outcome measures. Two patients had significantly better analgesia on more than one outcome measure on within-patient testing. One had better analgesia with DM. The other had better analgesia with placebo. Ten patients had no adverse effects on either dose of DM. Two patients withdrew during the first treatment period because of adverse effects (which included increased pain intensity), and 5 during the second period. Five patients continued with DM after the study for 1-3 months. No long-term clinical benefit was apparent in those who continued with open DM. Dextromethorphan at either 40.5 or 81 mg daily did not relieve neuropathic pain.