There are three categories of antithrombotic agents: drugs which prevent fibrin formation (the anticoagulants and defibrinating enzymes), drugs which prevent platelet adhesion or aggregation (the antiplatelet drugs), and thrombolytic drugs which induce fibrin degradation. Clinical studies have now led to a better understanding of the relative value of these drugs in different thrombotic disorders. In addition, knowledge of the mechanism of action of some of these drugs has recently been much advanced. The anticoagulant drugs in clinical use are heparin and the oral anticoagulants. Heparin is a potent inhibitor of several steps of the intrinsic coagulation pathway through its effect on a plasma cofactor, antithrombin III. Its action is immediate, but heparin must be given parenterally. Oral anticoagulants act more slowly, by reducing the hepatic synthesis of biologically active factors II, VII, IX and X, but can be given by mouth. Heparin is therefore most suitable for starting anticoagulant treatment, while oral anticoagulants are generally used for prolonged therapy. The value of the anticoagulants as antithrombotic agents has been best assessed by studying their effectiveness in preventing and treating venous thromboembolic disease. Oral anticoagulants have been repeatedly shown to prevent venous thrombosis and pulmonary embolism in patients at high risk of developing these complications. However, the increased risk of postoperative bleeding has prevented their widespread use for this purpose in surgical patients. Recently, the use of low doses of heparin, given subcutaneously before and after surgery, has been shown to markedly reduce the incidence of venous thrombosis and pulmonary embolism (including fatal pulmonary embolism) after major elective abdominal surgery, and to produce only a slight increase of postoperative bleeding. This represents a major advance in anticoagulant prophylaxis of venous thromboembolism in surgical patients. However, low dose heparin prophylaxis is relatively ineffective in patients having hip surgery, and has not been evaluated in patients having other types of orthopaedic surgery. There is direct evidence that anticoagulant therapy prevents death and recurrent embolism in patients who have developed pulmonary embolism, and considerable indirect evidence that it prevents pulmonary embolism (and death from pulmonary embolism) in patients who have venous thrombosis. The incidence of further venous thromboembolism or bleeding during treatment appears to be minimised when heparin is given by continuous intravenous infusion in a dose sufficient to produce a moderate, but not excessive, prolongation of a heparin-sensitive, in vitro coagulation test. The tests most commonly used to monitor heparin therapy was based on either the whole blood clotting time or the activated partial thromboplastin time. There is evidence that 3 to 6 months of oral anticoagulant treatment after initial heparin therapy results in a significantly reduced frequency of further thromboembolism, but that longer treatment is of little benefit except in patients with recurrent disease.