Antithrombin (AT) is a plasma-derived, single-chain glycoprotein with a molecular weight of 58 kDa. It is a serine protease inhibitor (serpin), sharing about 30% homology in amino acid sequence with other serpins. AT is a complex molecule with multiple biologically important properties. It is a potent anticoagulant that has been demonstrated to provide benefit in animal models and small cohorts of patients with coagulation disorders. AT also has remarkable anti-inflammatory properties, several of which result from its actions in the coagulation cascade. Activated coagulation proteases like activated factor X and thrombin contribute to inflammation; for instance, by the release of pro-inflammatory mediators. Inhibition of these proteases by AT prevents their specific interaction with cells and subsequent reactions. Anti-inflammatory properties of AT independent of coagulation involve direct interactions with cells leading to the release of, for instance, prostacyclin. Binding of AT to a recently identified cellular receptor, syndecan-4, leads to the interference with the intracellular signal induced by mediators like lipopolysaccharides and, thereby, to a down-modulation of the inflammatory response. AT has been shown to be effective in prospective and well-controlled small-scale studies of patients with inflammatory conditions, including sepsis. Although AT did not decrease overall patient mortality in a double-blind, placebo-controlled, phase III trial of patients with sepsis, it is important to note that AT improved the survival of individuals in this study not receiving heparin as a prophylactic regimen, which can be explained by the impaired interaction of AT with its cellular receptor in the presence of heparin, resulting in the reduction of the anti-inflammatory properties. Accordingly, the supplementation of AT without concomitant heparin may be beneficial in disorders with inflammatory characteristics, which has to be demonstrated in further clinical studies. Finally, recent results suggest that latent AT can induce apoptosis of endothelial cells by disrupting cell-matrix interactions. Further investigations will have to demonstrate whether latent and/or cleaved AT are physiological means to control angiogenesis. A potential prophylactic or therapeutic use as an anti-angiogenic and antitumor agent merits further exploration, including whether the growth of vessels in tumor tissues or close to tumors can be controlled by latent AT without affecting the formation of blood vessels during wound healing processes.