This systematic review of randomised or pseudorandomised trials aimed at summarising the effectiveness and safety of artemisinin drugs for treating severe falciparum malaria in adults and children. Survival was better with artemisinin drugs in 1.265 patients compared with 1.183 treated with quinine (OR: 0.68; 95% CI: 0.55-0.84). However, the difference is barely significant when only studies with adequate concealment of allocation at enrolment are included in the analysis (OR: 0.77; 95% CI: 0.61-0.98). In 1784 patients with cerebral malaria, mortality was also lower with artemisinin drugs overall (OR: 0.70; 95% CI: 0.55-0.90), but not significantly better than quinine in studies reporting adequate concealment of allocation. No difference in neurological sequelae has been demonstrated. Artemisinin drugs clear parasites from the blood faster than quinine. Adverse effects are similarly common with artemisinin drugs and quinine, although reporting varies between trials. There is no evidence from this review that any one artemisinin derivative is better than the others, but comparative studies are few, small and heterogeneous.