A study was conducted to test the hypotheses that antithrombin III (antithrombin) improves disseminated intravascular coagulation (DIC) when applied to DIC patients diagnosed by sensitive criteria and that the administration of high-dose antithrombin is a beneficial treatment for DIC. Twenty-three DIC patients diagnosed based on the Japanese Association for Acute Medicine (JAAM) DIC diagnostic criteria were treated with either high-dose (60 IU/kg/day) or low-dose (30 IU/kg/day) antithrombin concentrates for 3 days. The clinical conditions that cause DIC were restricted to systemic inflammatory response syndrome (SIRS) and sepsis. Data of antithrombin activity, platelet counts, coagulation and fibrinolytic markers, and DIC scores before antithrombin administration (day 0), on days 1 to 3, and on day 7 were retrospectively collected from computer-based records. Patients who met the JAAM DIC criteria were administered either high-dose (12 patients) or low-dose (11 patients) antithrombin. The patients' backgrounds and antithrombin activity (high dose, 51.5 +/- 14.5%; low dose, 62.6 +/- 19.3%; P = .153) on day 0 were identical in the 2 groups. The JAAM DIC score and prothrombin time ratio on day 7 significantly improved when compared with those on day 0. However, mortality at 28 days as well as interaction within the antithrombin doses administered showed no difference. There were also no differences in the time course of the platelet counts, coagulation and fibrinolytic markers, and DIC scores in the 2 groups. The authors conclude that the effects of antithrombin on prognosis and coagulation and fibrinolytic parameters are independent of the doses administered in patients with SIRS/sepsis-associated DIC.