The authors carried out a review of original safety data from three double-blind, placebo-controlled, randomised multicentre trials of Hypericum perforatum extract. The three trials had very similar protocols. The analysis included 594 outpatients suffering from mild to moderate depression (DSM-IV), treated with 3 × 300 mg/day H. perforatum extract (WS 5570, WS 5572, WS 5573) or placebo for 6 weeks. The risk ratios and risk differences vs. placebo for single and grouped adverse events were calculated for the polled data from the three trials. The data were analysed for relevant differences between H. perforatum extract and placebo and then compared to data from trials of mirtazapine and amitriptyline. For the polled data of the three placebo-controlled trials the incidence of any adverse events was comparable under H. perforatum and placebo. The adverse events typically associated with the administration of tricyclic antidepressants or selective serotonin reuptake inhibitors, such as sedation, anticholinergic effects, gastrointestinal disturbances and sexual dysfunction, were not observed in H. perforatum-treated patients. This analysis did not reveal any specific adverse effects of H. perforatum extract.