A randomized, open-label comparison of artesunate and quinine was conducted in 113 adults with clinically severe falciparum malaria in western Thailand. Mortality was 12% with artesunate and 22% with quinine treatment (relative risk, 0.53; 95% confidence interval, 0.23-1.26; P=.22). Multiple logistic regression analysis found admission plasma lactate level, Glasgow Coma Scale score, and total serum bilirubin level to be independent risk factors for death. Coma recovery and times to normalize plasma lactate levels were similar, but the parasite clearance time was much shorter among artesunate-treated patients (P=.019). Fewer patients became hypoglycemic during artesunate therapy (10%) than during quinine therapy (28%) (P=.03). Artesunate is at least as effective as quinine in the treatment of adults with severe malaria. Larger trials are required to determine whether mortality is reduced among patients treated with artesunate.