Seizures are poorly controlled in many people with epilepsy despite adequate current antiepileptic treatments. There is increasing interest in alternative therapies such as acupuncture; however, it remains unclear whether the existing evidence is rigorous enough to support the use of acupuncture.To determine the effectiveness and safety of acupuncture in people with epilepsy.We searched the Cochrane Epilepsy Group's Specialized Register (June 2005) and the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library Issue 3, 2005). We also searched MEDLINE, EMBASE, CINAHL, AMED, TCMLARS, China Biological Medicine Database, Chinese Acupuncture Trials Register, National Center for Complementary and Alternative Medicine, and National Institute of Health Clinical Trials Database from inception to June 2005. Reference lists from relevant trials were reviewed. No language restrictions were imposed.Randomised controlled trials evaluating any form of acupuncture involving people of any age with any type of epilepsy were included. Trials included were those comparing acupuncture with placebo, sham or no treatment; or comparing acupuncture plus other treatments with the same other treatments. Trials that only compared different acupuncture methods or compared acupuncture alone with other treatments were excluded.Two review authors independently extracted trial data and assessed trial quality using the Jadad score. Relative risk (RR) was used for binary data and weighted mean difference for continuous data, and 95% confidence intervals are given. Where possible, analyses were by intention to treat.Three small trials of varying methodological quality and with short follow up met the inclusion criteria. Two studied children in China and one studied adults in Norway. The two Chinese studies compared acupuncture plus Chinese herbs with Chinese herbs alone while the Norwegian study compared acupuncture with sham acupuncture. The two Chinese studies found that more children treated with acupuncture achieved 75% or greater reduction in seizure frequency (RR 1.52, 95% CI 1.12 to 2.05) and seizure duration (RR 2.38, 95% CI 1.13 to 5) with a significant 50% or greater reduction in seizure duration (RR 1.36, 95% CI 1.07 to 1.73). However, the two trials were of low quality without adequate description of randomisation method, concealment of randomisation or blinding. On the other hand, the higher quality Norwegian trial found that acupuncture did not improve the mean seizure frequency, seizure-free weeks, or quality of life in adults.The current evidence does not support acupuncture as a treatment for epilepsy. Much larger high quality clinical trials employing appropriate controls are needed.