New insights in heparin-induced thrombocytopenia

Comment on Rauova et al, page 131, and comment on Savi et al, page 139

Two papers in this issue of Blood, one by Rauova and colleagues and another by Savi and colleagues, go far in advancing our understanding of HIT.

To the clinician involved on a daily basis with the care of patients receiving unfractionated or low-molecular-weight heparin (UFH, LMWH), the most widely employed glycosaminoglycan (GAG) anticoagulants, heparin-induced thrombocytopenia (HIT) is a constant preoccupation. HIT is uncommon, but thrombocytopenia in hospitalized patients is very common and forces consideration of HIT as the explanation. The widely available enzyme immunoassay for the detection of etiologic heparin–platelet factor 4 (PF4) antibodies is too sensitive on the one hand and occasionally yields false-negative results on the other. Thus, diagnosis becomes a painstaking exercise involving (1) assessing whether the time to onset of thrombocytopenia and offset following drug discontinuation are consistent with the accepted time frame of HIT; (2) considering other causes of thrombocytopenia; and (3) pondering the meaning of the enzyme immunoassay results. The importance of this exercise cannot be overestimated—HIT-related thrombotic events are unpredictable and clinical miscalculation may lead to devastating consequences.

About a decade ago, Greinacher and colleagues¹  showed that the antigenicity of the GAG-PF4 complex inciting the development of immunoglobulin G (IgG) antibodies responsible for HIT depended on the molar ratio between the GAG and PF4, the composition of the GAG, and the length of the GAG (optimally 16 saccharides). Reilly and colleagues²  demonstrated that platelet FcγRIIA receptors were the target of the antigen-antibody complexes. Rauova and colleagues have now extended these findings in this issue of Blood. UFH formed stable ultra-large (greater than 670 kDa) complexes with PF4 at optimal molar ratio, LMWH formed mostly smaller complexes (44 kDa–120 kDa), and fondaparinux, a relatively smaller GAG with only 5 saccharides, formed no ultralarge complexes at all. Ultralarge complexes activated platelets (P-selectin surface expression, annexin V binding, and glycoprotein IIb/IIIa activation) in the presence of a murine monoclonal anti–heparin-PF4 antibody far more efficiently than small complexes. These findings fit neatly with clinical observations that UFH causes HIT more frequently than LMWH^3,4  and that fondaparinux appears not to cause HIT, even in orthopedic patients who are at highest risk for HIT.⁵ 

Savi and colleagues, in a blinded serologic study reported in this issue of Blood, compared the capacities of UFH and fondaparinux to stimulate platelets in the presence of 39 HIT sera. Studies were performed in laboratories expert in the less available but more specific functional assays (serotonin release, platelet agglutination, and platelet aggregation). UFH was positive in 79.8% of tests, whereas fondaparinux was positive in only 3.3% of tests. By flow cytometric techniques, UFH stimulated platelets in the presence of HIT serum in 85.7% of tests, causing glycoprotein IIb/IIIa complex activation, dense granule release, and negatively charged phospholipid expression, whereas fondaparinux did not (0% of tests).

Could fondaparinux become a treatment option in HIT? Our current choices are argatroban and lepirudin, synthetic direct anti-thrombins approved by the U.S. Food and Drug Adminstration for this indication. Both require continuous intravenous infusion and laboratory monitoring; lepirudin may rarely cause life-threatening anaphylaxis (especially after bolus doses and upon retreatment); and each is expensive. LMWH is not an option because of an unacceptably high cross-reactivity rate with UFH. Fondaparinux has the advantage of once-daily subcutaneous administration without need for monitoring in patients with adequate renal function, and it is far cheaper than argatroban and lepirudin. The absence of HIT in the fondaparinux orthopedic trials and the studies of Rauova et al and Savi et al are extremely encouraging. But we need a clinical trial of fondaparinux in patients with HIT to prove that the drug is safe and efficacious.