Basic research indicates that systemic local-anaesthetic-type drugs that block sodium channels are effective in pain due to nerve damage. These drugs were first used as analgesics in the 1950s and they are still commonly used to try to relieve neuropathic pain and incident pain caused by cancer. As they are potentially toxic, these drugs should not be used without proven effectiveness. For these reasons, a systematic review of randomized controlled trials of systemically administered local-anaesthetic-type drugs in chronic pain was performed. Main outcomes were pain relief or pain intensity difference and adverse effects. Twenty-one reports were found, and four publications were excluded. In the remaining 17 studies (450 patients), 10 used intravenous lignocaine, two used intranasal lignocaine, four used oral mexiletine and one used oral tocainide. The best documented effective dose of intravenous lignocaine was 5 mg/kg, and when infused over 30 min it was well tolerated. Intravenous lignocaine was effective in all four studies in non-cancer-related neuropathic pain. In migraine, lignocaine produced an inconsistent effect. Lignocaine was without effect in all three studies in cancer-related pain. Oral mexiletine showed some efficacy in all three studies in pain due to peripheral nerve damage, but lacked effect in the only study in central pain. Only minor dose-related adverse effects were reported in the 85 patients given mexiletine 225-750 mg. Local-anaesthetic-type drugs are effective in pain due to nerve damage, but there is little or no evidence to support their use in cancer-related pain. Copyright 1998 European Federation of Chapters of the International Association for the Study of Pain.