OBJECTIVE To evaluate the safety and efficacy of a natural platelet-activating factor receptor antagonist, BN 52021 (Ginkgolide B), in the treatment of patients with sepsis syndrome. METHODS Prospective, randomized, placebo-controlled, double-blind, phase III, multicenter clinical trial. METHODS Twenty-one academic medical center intensive care units in France. METHODS Two hundred sixty-two patients with sepsis syndrome who received standard supportive care and antimicrobial therapy, in addition to the administration of platelet-activating factor receptor antagonist or placebo. METHODS Patients received either a 120-mg dose of platelet-activating factor receptor antagonist intravenously every 12 hrs over a 4-day period or placebo. METHODS All patients were evaluated for 28-day, all-cause mortality. RESULTS The 28-day mortality rate was 51% for the placebo group and 42% for the platelet-activating factor receptor antagonist group (p = .17). However, the efficacy of platelet-activating factor receptor antagonist was significantly greater in patients with Gram-negative sepsis (test for interaction, p = .03). In a separate analysis of patients with and without Gram-negative sepsis, the 28-day mortality rate was 57% for the patients receiving placebo (30 deaths of 53 patients) and 33% for patients receiving platelet-activating factor receptor antagonist (22 deaths of 67 patients; p = .01). Platelet-activating factor receptor antagonist also significantly (p = .01) reduced the mortality rate among patients with Gram-negative sepsis who were in shock at entry into the study (mortality rate was 65% for placebo vs. 37% for platelet-activating factor receptor antagonist) and among patients > 60 yrs of age (mortality rate was 74% for placebo vs. 31% for platelet-activating factor receptor antagonist). A Cox proportional-hazards model identified five independent prognostic factors: a) adequacy of antibiotic therapy; b) severity of illness; c) renal failure; d) hematologic failure; and e) hepatic failure at study entry. When the Gram-negative sepsis population was stratified by age and these five prognostic factors were controlled for, the relative risk of death of the platelet-activating factor receptor antagonist group was 0.61 (0.34 to 1.08, 95% confidence interval; p = .09). This risk corresponds with an adjusted reduction in mortality rate of 39% for patients receiving platelet-activating factor receptor antagonist. No differences in mortality rates were found between the placebo and the platelet-activating factor receptor antagonist groups in the absence of Gram-negative sepsis. There were no differences in adverse events between the placebo and the treated groups. CONCLUSIONS The studied platelet-activating factor receptor antagonist (BN 52021) seems to be a safe and promising treatment for patients with severe Gram-negative sepsis.