To report the use of recombinant activated factor VII (rFVIIa) in controlling life-threatening bleeding episodes in patients with grades III and IV Dengue Hemorrhagic Fever (DHF), also known as Dengue Shock Syndrome. Fifteen patients (seven boys, eight girls), whose median age was 8 years, were enrolled in the study. They were divided into two groups. Group 1 included nine patients, mainly grade III, waiting for platelet concentrate, and group 2 included six patients, mainly grade IV, who had already received platelet concentrate with unresponsiveness. A single dose or repeated doses of 100 microg/kg rFVIIa was/were given at intervals of 4 h according to the bleeding symptoms. The median times from the onset of bleeding to rFVIIa initiation were 6.5 and 29.8 h in groups 1 and 2, respectively. Each patient received one to three doses. An effective response was found in eight patients (53.3%), including six patients in group 1 and two patients in group 2. They had complete cessation of bleeding without recurrence for 48 h. An ineffective response was found in seven patients (46.7%) including three patients in group 1 and four patients in group 2 for which the bleeding recurred (n = 2), temporarily slowed down (n = 3), continued (n = 1) or occurred at a new site (n = 1). These included three patients in profound shock 24-48 h before referral to comprehensive treatment centers, two patients receiving ibuprofen before hospitalization, one patient with extensive volume overloading, and one patient requiring surgical intervention to ligate the torn intercostal artery and vein. The platelet concentrate was promptly transfused to stop bleeding in patients with ineffective responses. The results revealed that the earlier initiation of rFVIIa in the mainly grade III DHF in group 1 yielded a higher effective response (66.7%) than the delayed initiation in the mainly grade IV DHF in group 2 (33.3%). Moreover, patients previously receiving ibuprofen or volume expander of low molecular weight dextran or urea-linked gelatin tended to have lower effective responses (28.6%) than patients without associated medication (75.0%). Ultimately, three of six patients with grade IV DHF died, while all nine patients with grade III DHF survived. Thus, the case-fatality rate in this study was 20%. No clinical evidence of thromboembolic complications was observed. rFVIIa seems to be effective in restoring hemostasis in a limited series of patients with Dengue Shock Syndrome exhibiting life-threatening bleeding episodes. Further study is warranted.