Heparin-Induced Thrombocytopenia: Molecular Pathogenesis

Abstract

IntroductionHeparin-induced thrombocytopenia (HIT), a relatively common complication of heparin therapy, is characterized by an unexpected fall in platelet count occurring 5 days or more after the initiation of treatment.1,2 Patients who manifest only the thrombocytopenia rarely experience severe bleeding or other major adverse effects.3 Often, however, the thrombocytopenia is accompanied by arterial and/or venous thrombosis and thromboembolism, a complication which can be devastating.4 HIT, with or without thrombosis, is probably the most common cause of immunologically-mediated, drug-induced thrombocytopenia. HIT usually occurs in patients given standard therapeutic doses of heparin, especially of bovine origin.5 However, it has been reported following low-dose, subcutaneous heparin6 and in patients exposed to heparin “flushes” used to maintain the patency of intravenous lines.7 Even minute quantities of heparin released from heparin-bonded catheters appear to be capable of causing the disorder in previously sensitized patients.8 HIT has been induced by low molecular weight heparin (LMWH)9,10 and other heparin-like polysaccharides.11,12 HIT differs from most other forms of drug-induced immune thrombocytopenia in that the responsible antibodies activate platelets in the presence of pharmacologic doses of heparin (0.1-1 U/ml), rather than merely binding to platelets to promote their destruction in the reticuloendothelial system.3,13,14 The observation that heparin-induced platelet activation can be blocked in vitro by monoclonal antibodies specific for the platelet Fc receptor (FcγRIIA)15 suggests that antibodies associated with HIT somehow interact with heparin to form platelet-activating immune complexes. Attempts to demonstrate such complexes and their binding to intact platelets, however, have generally yielded negative or equivocal results.14,16,17 The association of HIT with thrombosis and disseminated intravascular coagulation (DIC) suggests that heparin-induced antibodies might react with endothelial cells (EC). Binding of IgG from the serum of patients with HIT to cultured human umbilical vein endothelial cells (HUVEC) observed in one study,18 but not confirmed in another,19 left open the question of whether immune-mediated vascular damage is important in pathogenesis.In 1992, Amiral and coworkers obtained evidence that antibodies associated with HIT are specific for complexes of heparin and platelet factor 4 (PF4), a heparin-binding protein stored in platelet α granules.20 These findings were subsequently confirmed by several groups, including ours.21-23 We also provided the first direct evidence that HIT antibodies bind to HUVEC in the presence of PF4, but not in its absence, thus, suggesting that the antibodies also recognize PF4 complexed with the heparan sulfate molecules displayed on HUVEC.21 Furthermore, Greinacher et al22 showed, by adsorption/elution of total IgG derived from HIT patients on PF4-coated EC, that the eluted antibodies also caused heparin-dependent platelet activation, thus, indicating that they recognize epitope(s) present on both PF4-heparin and PF4-heparan complexes.