Disseminated intravascular coagulation is a fundamental pathogenetic mechanism which complicates a wide variety of diseases. The clinical and hematologic problem varies so greatly from one disease process to another and from one patient to another with the same disease that no standard therapy is possible. Each patient must be handled individually. The fact that disseminated intravascular coagulation is an intermediary mechanism of disease indicates that its best prevention or treatment would come from prevention or treatment of the underlying disease process. Since this is not possible at present, treatment directed at the coagulation and fibrinolytic systems is indicated. Evidence of disseminated intravascular coagulation in any disease is obtained from three sources: (1) pathologic anatomy, (2) alterations in the hemostatic mechanism and (3) clinical manifestations. All of these must be understood before therapy is undertaken. Anticoagulants (heparin and Dicumarol) and fibrinolysin activators (streptokinase and urokinase) constitute the therapeutic agents that interfere directly with the coagulation process. Replacement therapy (whole blood, plasma and fibrinogen) and fibrinolysin inhibitors (EACA and Trasylol) are indicated when clotting has ceased and activation of the fibrinolytic system is the major problem. In most diseases clotting and fibrinolysis are superimposed on each other. Actual clinical experience with the use of these agents in disseminated intravascular coagulation is still in a very early stage of development, but promising results have already been obtained. One of the important contributions of these early studies is the demonstration of a rapid return to normal of the clotting factors after heparin has been used. This is valuable information in itself because it constitutes further evidence that disseminated intravascular coagulation is occurring in the patient and in the disease process. Heparin therapy in single cases has apparently been effective in stopping bleeding and in prolonging life. The best clinical and laboratory responses have been achieved in diseases with subacute or recurrent intravascular clotting, including thrombohemolytic thrombocytopenic purpura, hemolytic-uremic syndrome, purpura fulminans and Philippine hemorrhagic fever. Symptomatic response has also been described in cases of cancer, leukemia, cirrhosis of the liver and giant hemangioma. The fears of the dangers of the use of anticoagulants in these syndromes have been exaggerated. The cases in which serious hemorrhage could in any way be attributed to the anticoagulant are practically nil. Attention has been drawn to the risk of maintaining intravascular thrombi by epsilon aminocaproic acid (EACA) in patients who have a continuing intravascular clotting even though the fibrinolytic system has been activated. Several cases of bilateral renal cortical necrosis have been reported after administration of fibrinogen and EACA. The combined use of heparin and EACA is probably safer. EACA is very useful in reducing the hemorrhage from local bleeding sites, as in prostatic resection, whether or not there is a systemic abnormality of the hemostatic process. Many of the reported good results with EACA in patients with disseminated intravascular coagulation must be viewed in the light of the local effect on bleeding sites. In many of the cases reviewed, multiple therapeutic agents have been used, and precise evaluation of the effectiveness of any single agent is obscured. With the exception of heparin therapy following pulmonary embolism, the treated cases are too few to al ow statistical evaluation of therapy. More careful clinical studies are needed to establish the effectiveness of these agents on a statistical basis.