Pathophysiology of Heparin-Induced Thrombocytopenia: Clinical and Diagnostic Implications—A Review Open Access

Among the known side effects of heparin therapy, thrombocytopenia is without doubt the most frequent and dangerous. There are 2 types of heparin-induced thrombocytopenia (HIT). Heparin-induced thrombocytopenia I is characterized by a transitory and asymptomatic reduction in the platelet count, rarely below 100 × 10⁹/L, that resolves spontaneously and does not require removal of the drug. The origin of HIT I is not yet completely known, but is thought to be related to a phenomenon of heparin-induced platelet clumping.1–3 No immunologic components are involved in HIT I, and pathologic manifestations are rare.

Heparin-induced thrombocytopenia II has an immunologic origin4,5 and is characterized by a significant reduction in platelets (>30%), generally after the fifth day of therapy; in the case of previous exposure to heparin, thrombocytopenia may appear earlier.6 The thrombocytopenia usually resolves 5 to 15 days after heparin has been removed, but in some cases it may take months.6–8 The pathophysiologic manifestations of HIT II are complex and involve thrombosis at arterial, venous, and microvascular sites.

The true incidence of HIT II is not well defined because reported studies are mostly retrospective and differ regarding the characteristics of the patients considered, type of heparin administered, dosage, route of administration, duration of therapy, definition of thrombocytopenia, and laboratory tests employed for diagnostic confirmation (Table 1).7,9,10 With reference to prospective studies in which the diagnosis was clinically based, there are important differences in the definition of thrombocytopenia. In some studies, the threshold value is 150 × 10⁹/L,11–17 while in others it is 100 × 10⁹/L18–30; other investigators use the percent decrease as their reference.31 A relationship between the incidence of HIT II (defined only on a clinical basis), dosage, and type of unfractionated heparin used emerged from a study by Warkentin.32 The incidence was about 5% for therapeutic dosages of bovine heparin and 1% for porcine heparin, while it was less than 1% with prophylactic dosages of porcine heparin. In this series, the incidence of secondary thrombotic complications was about 20%. In a later review of prospective clinical trials,7 the incidence of HIT II varied from 1% to 30% in patients treated with high dosages of intravenous heparin, while it was less than 2% in patients administered low dosages of subcutaneous heparin.7 

Schmitt and Adelman10 reviewed 23 randomized or cohort prospective studies for a total of 2160 patients in order to evaluate the impact of the various methodologic characteristics, such as HIT II definition, frequency with which platelet count was verified, and diagnostic criteria. This analysis confirmed that the incidence of HIT is overestimated in studies that do not include a “repeatedly abnormal platelet count.” The cumulative incidence of HIT II in studies that employed “a reproducibly lowered platelet count” was 2.9% for bovine heparin and 1% for porcine heparin, and 1.7% for intravenous administration and 0% for subcutaneous administration. Even if this trend does not reach statistical significance, it speaks in favor of porcine heparin and subcutaneous administration of low dosages.

Our retrospective study disclosed a higher incidence.33 Independent of the route of administration, 6% of the patients had a clinical score suggestive of HIT II, with a 30% incidence of thrombotic complications and a 30% mortality rate; these values are in line with other published reports.6,8,34 However, using more selective clinical criteria, the percent incidence lowered to 3% and the diagnosis was confirmed by the presence of anti–heparin-platelet factor 4 (anti-H-PF4) antibodies in only a fraction of patients.33 

On the other hand, Kappers-Klunne et al35 reported a particularly low (0.3%) HIT incidence in 558 cardiologic and neurologic patients treated with intravenous heparin. In this study, both functional and immunologic tests were used for laboratory confirmation of the clinical diagnosis. Anecdotal reports36,37 describe HIT II induced by low-molecular-weight heparin (LMWH), but one clinical study indicated that its use is associated with a lesser incidence of thrombocytopenic and thrombotic complications than heparin.38 

A recent double-blind randomized study compared subcutaneous heparin with LMWH in 655 patients undergoing orthopedic surgery38; the clinical diagnosis of HIT II was confirmed by means of the radioactive carbon (¹⁴C)-serotonin-release assay (SRA). Heparin-induced thrombocytopenia II was documented in 2.7% of the patients treated with subcutaneous heparin and in none of the patients receiving LMWH (P = .0018). Thrombotic complications were also more frequent in the former (88.9%) than in the latter (17.8%) group (P < .001). In a subgroup of patients, independent of the presence of HIT II, more heparin-treated than LMWH-treated patients had a positive functional test (7.8% vs 2.2%, P = .02); thrombotic episodes, however, were more frequent in the patients who developed HIT II than in those with only a positive functional test. Thus, the frequency of laboratory-confirmed HIT II seems to be about 2% in patients receiving heparin, while it is much lower in those who receive LMWHs.

An immunologic basis of HIT II was first advocated by Rhodes et al,39 who showed that the immunoglobulin (Ig) G fraction from the serum of patients with HIT caused in vitro platelet aggregation in the presence of therapeutic concentrations of heparin. It has been reported that immunoglobulin-heparin complexes form only in the presence of platelets (Table 2).40 The proaggregating effect of heparin depends on the degree of sulfation and the molecular weight,41–43 and is mediated by the release of substances from platelet α-granules.44 Several platelet proteins/chemokines were proposed as the putative receptors of heparin-dependent antibodies,45 and PF4 was identified as the main cofactor.46,47 The ratio of heparin to PF4 is critical for the constitution of the multimolecular antigenic complex, with an optimal heparin-PF4 ratio ranging from 1:4 to 1:6.6,43,48–50 The antibody is not exclusively specific for the heparin-PF4 (H-PF4) complex, but also marks complexes of PF4 and other glycosaminoglycans.41–43,51 

The Figure shows a diagrammatic representation of a modified version of the currently accepted mechanism of action and pathophysiology of HIT (Table 2). At therapeutic concentrations ranging from 0.1 to 1.0 U/mL, heparin displaces PF4 from endothelial heparan sulfate or releases it directly from the platelets. Numerous PF4 molecules bind to heparin components, and the complex becomes immunogenic. As illustrated in the Figure, the immune complexes made up of anti-H-PF4 antibodies leading to the generation of 3 groups of antibodies (mainly IgG) activate the platelets and provoke an immune-mediated endothelial lesion4,6,47,51,52 with thrombocytopenia, thrombosis, or both. The IgG anti-H-PF4 immune complex activates the platelets through the bond with the FcγRIIa (CD32) receptor,53 whose platelet surface expression ranges from 700 to 4000 binding sites and is further increased by the immune complex bond.54,55 Platelet activation is blocked by both the monoclonal antibody (IV.3) specific for the FcγRIIa receptor50,56 and the F(ab′)₂ fractions from patients with HIT II.56,57 The Arg/His polymorphism at position 131 of the FcγRIIa receptor influences platelet reactivity to the immune complexes55; in particular, the His/His phenotype is more reactive to the IgG₂ isotype. Nonetheless, while some studies have demonstrated a greater prevalence of HIT II and thrombotic complications in subjects with the His/His phenotype,58 others have not confirmed these findings.59 Other data are consistent with the hypothesis that H-PF4 complexes bind directly to platelets, and these complexes are the target for the F(ab′)₂ fraction of the antibody.60 

How the anti-H-PF4 antibodies cause thrombosis is not clear. In general, IgG₂-isotype antiheparin antibodies are not particularly more frequent than the other subclasses in patients with HIT II,61 and IgM and IgA, which are not able to bind to FcγRIIa receptors, are also present in a significant percentage of these patients.62–65 This finding suggests that the mechanism of platelet activation may occur independent of the FcγRIIa receptor for IgG. Moreover, the antibody isotype tends to modify in relation to the duration of the treatment.49,61–64 The antibodies are still detectable in patient serum for about 4 to 6 weeks, and cases of antibody persistence for longer periods of time have been described.6,33 Although the thrombotic complications in HIT syndrome are well described, only limited data have become available on the inflammatory components in this disease.

We have proposed a functional heterogeneity of anti-H-PF4 antibodies, based on the fact that heparin is a heterogeneous mixture of sulfated mucopolysaccharides with molecular, structural, and physical heterogeneity.66 Thus, heparins likely form multiple complexes with PF4, and depending on the nature of this interaction, the allosteric modifications in PF4 leading to a neoantigen formation may also vary. To characterize the anti-H-PF4 antibody in terms of functional heterogeneity, we obtained IgG fractions from the serum of patients with HIT II utilizing ammonium sulfate precipitation (ASP) and H-PF4-sepharose 4B affinity chromatography methods.67 With affinity purification, 2 major components, peaks 1 and 2, with high anti-H-PF4 antibody titers were eluted (purity was established by sodium dodecyl sulfate-polyacrylamide gel electrophoresis). While peak 1 (despite having a high anti-H-PF4 antibody titer) did not induce serotonin release from platelets in a heparin-dependent manner, peak 2 and the IgGs obtained with the ASP method exhibited a strong and concentration-dependent activation in the presence or absence of heparin (as well as LMWHs) (Table 3). These data suggest the generation of “superactive” HIT antibodies capable of activating platelets without heparin. The anti-H-PF4 antibody titers of peak 1, peak 2, and the ASP-IgG as measured by the heparin-induced platelet aggregation test–enzyme-linked immunosorbent assay (ELISA) (Stago, Asnieres, France) were similar.

To rule out the possible presence of a heparin contamination in these IgG preparations (and thus to confirm antibody activity independent of heparin), heparinase digestion, the use of an ion-exchange resin (Heparsorb), and dialysis against 1× phosphate-buffered saline were performed on the HIT sera, peak 2, and IgG-ASP.67 None of these treatments resulted in a significant decrease in the activation of platelets.67 These observations underscore the complex pathophysiology of HIT syndrome and suggest that there may be an HIT antibody active in a non-heparin-dependent manner.

Because the pathophysiology of HIT II involves the activation of platelets, endothelial cells, and leukocytes, it is reasonable to assume that cellular activation products, such as soluble selectins, cellular adhesion molecules, or both, would be increased in HIT. Our studies showed that selectin levels were markedly elevated in HIT II patients (Table 4). Treatment of these patients with the direct thrombin inhibitor argatroban (Novastan, Texas Biotechnology Corp, Houston, Tex) was associated with some decrease in the level of selectins at 24 and 72 hours posttreatment. A similar decrease of soluble P-selectin levels was also noted when HIT patients were treated with other direct thrombin agents (eg, hirudin and hirulog) as alternative anticoagulation therapies.68 Since thrombin plays an important role in the activation of platelets, resulting in microparticle formation, the selectin-level down-regulation may be related to the inhibition of thrombin generation. In addition, we have observed an increase in the intracellular adhesion molecules and the vascular cell adhesion molecules in HIT patients.68 

In an experimental mouse model, the formation of autoantibodies against the H-PF4 complex produced thrombocytopenia but not thrombosis.65 From a pathogenic point of view, it is likely that a state of platelet and endothelial cell preactivation and probably other unidentified factors contribute to the thrombotic phenomena.6,52,69,70 

In HIT II, the onset of thrombocytopenia appears to be independent of the type of heparin, dosage, and route of administration.18 The entity of thrombocytopenia usually varies from 50 to 100 × 10⁹/L, but severe cases are frequent (Table 5).7,71 There is no gender predominance,9 although elderly patients undergoing postsurgical prophylaxis or treatment for deep vein thrombosis,71,72 in particular orthopedic and cardiovascular surgery,8 seem to be at higher risk. In more than 60% of the cases, other concomitant prothrombotic factors exist, such as diabetes, neoplasm, cardiac insufficiency, systemic lupus erythematosus, antiphospholipid syndrome, infection, and trauma. Besides thrombocytopenia, cutaneous allergic manifestations and skin necrosis may be present.72 

Despite the thrombocytopenia, hemorrhagic events are not frequent, while the major clinical complication is thrombosis. Both arterial and venous thrombosis can complicate the course of HIT II. A worsening of the thrombosis or a new thromboembolic complication necessitates the initial heparin treatment.8,9,32,73 The thrombotic complications may appear even in the absence of thrombocytopenia.74 Arterial thrombosis was the first reported event to be associated with HIT39,75; nonetheless, today arterial and venous thrombotic complications are commonly found in HIT patients.76 Arterial thrombosis seems to be more frequent in patients with cardiovascular disease,34,71 and venous complications are found more often in patients undergoing postsurgical prophylaxis.8,9,34,71 The most common arterial complications are thromboses of the large vessels with gangrene and limb amputation, stroke, myocardial infarction, and cardiac thrombosis.4,5,32,34,71 Venous complications are deep vein thrombosis, pulmonary embolism, thrombosis of the cerebral venous sinus, and closure of arterial-venous fistula in dialyzed patients; disseminated intravascular coagulation and hemorrhagic adrenal necrosis have been documented occasionally.4–7,32 

It is commonly believed that HIT II is underdiagnosed. During heparin therapy, platelet counts must be checked regularly, at least twice weekly, especially in patients receiving treatment for more than 4 days, in those who show resistance to heparin, or in patients who have treatment-related skin manifestations (Table 6). Once thrombocytopenia is confirmed, the diagnosis of HIT II should be formulated on the basis of clinical criteria and the in vitro demonstration of heparin-dependent antiplatelet antibodies.5,7 Nonetheless, the diagnosis is still only clinically based (associated with negative laboratory results) in more than 20% of cases.76 To evaluate the clinical probability of HIT II, various scoring systems have been proposed based on the presence of thrombocytopenia, recovery following drug suspension, onset of thrombotic or cutaneous complications, and the exclusion of other causes of thrombocytopenia.77,78 A score less than 3 is not associated with a diagnosis of HIT II. Heparin-induced thrombocytopenia II is possible with a score from 4 to 6, and a score greater than 6 is highly probable for HIT II.78 

Among the functional laboratory tests, the SRA is the reference procedure.77 This test is based on the capacity of heparin-dependent HIT antibodies to induce the release of ¹⁴C-serotonin from platelets. Serum from a patient suspected of having HIT II is incubated with therapeutic heparin concentrations (0.1–1.0 U/mL) and washed donor platelets labeled with ¹⁴C-serotonin. If heparin-dependent antibodies are present, platelets are activated and the labeled serotonin is released. In the presence of high heparin concentrations (10–100 U/mL) the release is inhibited. This method has several disadvantages in that it requires the use of radioactive material, the result is highly dependent on the characteristics or reactivity of the donor platelets,79,80 and it is time-consuming.

The platelet aggregation test, which utilizes a principle similar to that of the SRA, is able to furnish quicker results.56 This test measures platelet aggregation induced by the HIT patient serum in the presence of therapeutic concentrations of heparin. While this method is widely used due to its relative simplicity, the results vary considerably more than those reported for SRA in relation to the different heparin concentrations and donor platelet variability.5,7,32,80,81 The overall sensitivity of this test is less than that of the SRA.80,81 

The heparin-induced platelet aggregation test on micro-ELISA plates demonstrates greater reliability and correlation with the SRA than the platelet aggregation test.82 This test is based on the visual evaluation of the aggregation of washed platelets from different donors in the presence of heparin utilizing a magnetically shaken microplate.

Regardless of the functional method used to detect HIT antibodies, the selection of the donor platelets is crucial.79,80 Under the most optimal conditions, the sensitivity of the aggregation and the SRA methods can reach 88% and 94%, respectively.4,5,7,32,83 The typical response, however, is 50% to 60% sensitivity of these assays.

Recently, other functional tests have been suggested, such as the bioluminescent adenosine nucleotide-release assay84 or the binding of annexin V to platelet membrane anionic phospholipids utilizing flow cytometry.85 As these tests also employ donor platelets, they too would be affected by platelet variability. Other flow cytometric assays have been developed using small volumes of patient platelets or whole blood.85,86 This approach could provide a major advantage over all other tests described if sensitivity and specificity to HIT are proven. Another recently described test employs a solid-phase adherence method to demonstrate the presence of heparin-dependent antibody.87 

Following the demonstration of antibodies against the H-PF4 complex in the serum of patients with HIT II,46 the ELISA technique for the detection of these antibodies was introduced.46,49,50 Patient serum is incubated with the H-PF4 complex, and the presence of antibodies is detected with a secondary antibody conjugated with peroxidase or alkaline phosphatase. The ELISA showed a good correlation with the SRA procedure,78,81,88 but comparison with the aggregation method was less reliable.78,81,89 The ELISA method is characterized by greater sensitivity and reproducibility than the functional tests,88 and this procedure is technically easier to perform. However, the ELISA has demonstrated questionable specificity, in that anti-H-PF4 antibodies are detected in heparin-treated patients who did not present thrombocytopenia and in most patients undergoing heart surgery.31,35,51,62,64,90–92 More importantly, this test was negative in some patients with HIT confirmed clinically or by positive functional tests.78,81,88 It has been suggested that HIT II only occurs with high antibody titers and after persistent exposure to heparin,64,90,93 and also that antigens different from the H-PF4 complex can be involved in the pathogenesis.94 In addition, assays from different manufacturers have different sensitivities and specificities.95 

In general, however, aside from the varying sensitivity levels of the methods and the lack of standardization, ELISA tests have proven to be predictive of the diagnosis of HIT II.64,90 Therefore, especially in the absence of a highly suggestive clinical picture, it appears appropriate to support the clinical diagnosis with a functional test, as well as with measurement of the anti-H-PF4 antibody titer by ELISA.48,79,82,87–89,96 

The best therapeutic strategy for patients with HIT II is not established, but reasonable guidelines have a wide consensus (Table 7). If HIT II is clinically probable, heparin therapy must be discontinued immediately, even in the absence of a confirmatory laboratory test. Platelet transfusion is contraindicated because it may worsen the thrombotic picture. Anticoagulant therapy with vitamin K antagonists should be initiated 3 or 5 days after heparin suspension, when platelet counts are increasing, but preferably before resolution of the thrombocytopenia to avoid potential worsening of the thrombotic picture.6,97 However, heparin discontinuation alone and substitution with dicumaroids do not prevent the onset of severe thrombotic complications in nearly 50% of affected patients.8,34,97 On the basis of these disappointing results, new approaches have been proposed that have included the use of LMWH, heparinoids, anticoagulating agents such as ancrod, prostaglandin (Iloprost), antiplatelet drugs, and thrombin inhibitors (argatroban and hirudin).

Low-molecular-weight heparins, heparinoids, ancrod, argatroban, and hirudin have been used in a significant number of patients.4–6,9,75,98–100 A small clinical trial was conducted with ancrod, a viper-derived venom with anticoagulant action that showed efficacy.101 Other reports describe the use of plasmapheresis to remove immune complexes,102 or high doses of immunoglobulin alone103 or associated with LMWH and a heparinoid.104 Also indicated for use in patients with HIT II are thrombolytic agents; the insertion of filters in the inferior vein cava; and, in the case of arterial thrombosis, with limb ischemia, embolectomy, or thrombectomy.4–6,92 

The rationale for the use of LMWHs in HIT II resided in the diminished interaction with PF4 of these heparins with decreased molecular weight and degree of sulfation.51,98,105 However, the cross-reactivity with heparin-induced antibodies in vitro was shown to range from 60% to 100%.4,51,81,97,106,107 Low-molecular-weight heparins should not be administered to patients with heparin antibody unless the absence of cross-reactivity has been demonstrated by an in vitro test.6 Nonetheless, some reports describe cases in which the use of LMWH was efficacious in controlling HIT even though cross-reactivity with heparin had been evidenced.104,108 

In a study from our group, 2 synthetic pentasaccharides (SR90107A/Org31540 and SanOrg34006), which are in clinical development for the prophylaxis of postsurgical deep vein thrombosis, were tested in comparison to heparin and an LMWH (enoxaparin) for their relative platelet activation potential in HIT assays.109 Sera from patients with HIT II (n = 25), validated for heparin-dependent aggregation responses, and antibodies purified by H-PF4-sepharose column separation were used to study the effects of the 4 drugs using platelet aggregation. At comparable concentrations, heparin and enoxaparin consistently produced platelet activation (Table 8), whereas both pentasaccharides failed to produce a response at concentrations up to 100 U/mL (∼50 μmol/L). Similarly, in the SRA and flow cytometric assays, both heparin and enoxaparin produced positive responses, whereas the 2 pentasaccharides consistently failed to produce any effect.

We have further shown that in patients from a clinical trial substudy in which pentasaccharide was administered for the prophylaxis of deep vein thrombosis after hip surgery, no anti-H-PF4 antibody was detected during the treatment period (n = 10). However, in a comparable study with enoxaparin, 6 of 20 patients without clinical thrombocytopenia showed a positive anti-H-PF4 antibody titer. The observations from these studies suggest that the pentasaccharides with highly selective anti-Xa activity are devoid of generating anti-H-PF4 antibody, do not produce HIT responses in the presence of antibody, and may inhibit active HIT antibody platelet activation.109 

The major reported experiences of HIT treatment concern danaparoid sodium (Org10172, danaparoid, Lomoparan; Organon, Oss, The Netherlands), a mixture containing heparan sulfate (85%), dermatan sulfate (10%), and chondroitin sulfate (5%), whose cross-reactivity with heparin in vitro is less than 10%.9,33,48 More than 600 patients with HIT II have been treated successfully with this drug, with a remarkable reduction in mortality due to thrombotic complications and in overall mortality in patients treated early.9,75,104,110,111 However, cases of failure of treatment106,112 or of danaparoid-induced fatal thrombotic thrombocytopenia113 have also been reported. In particular, treatment with danaparoid resolved thrombocytopenia in 91% of cases and significantly reduced mortality due to thrombotic complications of HIT from 28% to 5%, but did not reduce total mortality, which was 20%.9,76 Danaparoid has been approved for the treatment of HIT II in New Zealand, Denmark, Luxembourg, Belgium, and Portugal.

Direct thrombin inhibitors are indicated for the treatment of HIT II. The first large-scale clinical trial of HIT patients treated with a thrombin inhibitor used argatroban, a thrombin inhibitor based on the structure of L-arginine.114 A number of patients with HIT II–associated thrombosis requiring angioplasty were also successfully treated with argatroban.100,115 Hirudin, another thrombin inhibitor, was evaluated for use in the treatment of HIT II (mostly in Germany), and it was found to be efficacious.116,117 

A further theoretical possibility, especially for patients with severe thrombotic complications refractory to thrombin inhibitor treatment alone, is the use of antiplatelet agents.4–6 It has recently been demonstrated by several in vitro studies that antagonists of GPIIb/IIIa (Abciximab, Tirofiban, and Eptifibatide) are able to inhibit platelet aggregation induced by the serum of patients with HIT II.86,118–121 The in vitro inhibition by GPIIb/IIIa inhibitors was more effective than inhibition of platelet activation by aspirin, as shown by SRA, platelet aggregation, and flow cytometry assays.119 The clinical usefulness of this treatment, combined with low-dose antithrombin agents, has shown preliminary beneficial results.118,119 

Recently, we proposed that the prevalence of HIT antibodies in patients treated with immunosuppressive agents (such as cyclosporine) would be lower than in nontreated patients.122 Cyclosporine is used to suppress the immune systems of transplant recipients to prevent the production of antibodies against the foreign major histocompatibility factor of the donor organ, thus reducing the incidence of rejection. In testing the anti-H-PF4 antibody levels in cardiac surgery patients (n = 48) and cardiac transplant patients (n = 30), 23% of the cardiac surgery patients had positive antibody titers in contrast to 10% of transplant patients (Table 9).122 Of the positive cases, 6.3% of the cardiac surgery patients were positive by SRA, but none of the transplant patients were positive by SRA. Patients with rheumatoid arthritis (n = 9) and antiphospholipid syndrome (n = 21), which was treated with heparin and immunosuppressive therapy, did not exhibit a positive anti-H-PF4 antibody titer. These observations suggest that patients at high risk may be prophylactically treated with mild immunosuppression prior to heparinization to minimize the risk of HIT II. Clinical investigations are needed to validate this hypothesis.

The pathophysiology of HIT is now known to be mediated by antibodies to the anti-H-PF4 complex. These antibodies represent a heterogeneous group of IgG, IgA, and IgM antibodies that are generated in response to the neoepitope formed by the complex formation of heparin and PF4. Their functional form is capable of interacting with Fc receptors to activate platelets and endothelial cells. Cytokines are generated during this process as well. Activated platelets are consumed to form localized thrombi, resulting in the white clot syndrome. Inflammation plays an important role in the overall pathophysiology of HIT, and the elevated circulating levels of inflammatory markers have been found in patients with HIT.

The clinical diagnosis of this complex syndrome is based on platelet counts and, in extreme cases, the identification of characteristic purplish lesions on the skin. The laboratory diagnosis of HIT can be accomplished by using immunologic, platelet aggregation, and serotonin-release assays; however, the diagnostic efficacy of these tests is variable.

Patients with HIT can receive alternative anticoagulation therapy with several different drugs; however, only antithrombin drugs, such as argatroban and hirudin, are approved for this indication. Ancrod, a snake venom; danaparoid, a depolymerized mixture of glycosaminoglycans; and nonheparin glycosaminoglycans (dermatans) have also been used with success for alternative anticoagulation therapy. Plasmapheresis has been used for the management of HIT as well. Currently, synthetic pentasaccharides are being developed clinically, and the initial data are strongly suggestive of antithrombotic efficacy without any heparin antibody–related complications. Low-molecular-weight heparins are not indicated for the management of these patients. It has been shown that the activation of platelets during the acute HIT syndrome is not fully controllable by anticoagulant medications. Antiplatelet drugs, especially GPIIb/IIIa receptor antagonists, have been found to produce therapeutic effects and control the platelet-mediated pathophysiologic mechanisms more effectively than thrombin inhibitors.

Although we have made progress in better understanding the pathophysiology of HIT and have found better therapeutic options, there remain unsettled diagnostic and treatment issues that need to be addressed.