Heparin-Induced Thrombocytopenia

Heparin-induced thrombocytopenia (HIT) is a prothrombotic disorder caused by antibodies to complexes of platelet factor 4 (PF4) and heparin.[6] Rauova and colleagues reported that heparin and PF4 form stable, ultralarge (>670 kDa) complexes (ULCs) composed of multiple PF4 tetramers arrayed in a lattice with several molecules of unfractionated heparin. Far fewer ULCs were formed when low-molecular weight heparin was used, and none were formed with the factor Xa inhibitor fondaparinux.[7]

Zheng and colleagues reported that healthy humans possess preexisting inactive/tolerant PF4/heparin-specific B cells, and that breakdown of tolerance can lead to production of PF4/heparin-specific antibodies.[8] These HIT antibodies can be found in plasma in more than 90% of patients with the clinical diagnosis of HIT.[6] However, HIT antibodies are also present in many patients (especially patients undergoing cardiac procedures) who have been exposed to heparin but who do not have clinical manifestations of HIT.

The antibodies bind to the PF4-heparin complexes on the platelet surface and induce platelet activation by cross-linking FcγIIA receptors.[9, 10, 11] The activated platelets increase the release and surface expression of PF4, creating a positive feedback loop in which further release of PF4 promotes further platelet activation.[12]

Alternatively, HIT antibodies may recognize PF4 bound to platelet chondroitin sulfate. These antibodies activate platelets even in the absence of heparin, thus explaining delayed-onset HIT, persistent HIT (in which recovery takes several weeks), spontaneous HIT syndrome (which resembles HIT clinically and serologically but occurs without proximate heparin exposure), and fondaparinux-associated HIT.[13]

Platelet activation results in the release of procoagulant platelet microparticles, platelet consumption, and thrombocytopenia. Marked generation of thrombin, activation of monocytes and other inflammatory cells, and endothelial injury and activation follow, producing the characteristic venous and arterial thromboses of HIT.

Heparin-induced thrombocytopenia (HIT) is caused by antibodies that bind to complexes of heparin and platelet factor 4 (PF4), activating the platelets and promoting a prothrombotic state. HIT is more frequently encountered with unfractionated heparin (UFH) than with low molecular weight heparin (LMWH).[14]

The risk of HIT is highest with prolonged use of heparin for postoperative thrombophylaxis. However, case studies have also demonstrated the possibility of developing HIT with minimal heparin exposure via intravascular flushes to maintain the patency of indwelling arterial or venous catheters.[15, 16, 17, 18, 19]

Fondaparinux is a synthetic pentasaccharide that catalyzes the inhibition of factor Xa (but not thrombin) by antithrombin, and thus inhibits thrombin generation. A study suggested that fondaparinux may be associated with formation of anti–PF4/heparin antibodies but, in contrast to LMWH, is unlikely to cause HIT because of the poor reactivity of antibodies against PF4/fondaparinux.[20]

In the United States, approximately 12 million individuals, or one third of hospitalized patients, have some heparin exposure yearly. A study by Smythe and colleagues estimated the frequency of heparin-induced thrombocytopenia (HIT) to be 0.76% in patients receiving therapeutic doses of intravenous unfractionated heparin (UFH) and less than 0.1% in patients receiving antithrombotic prophylaxis with subcutaneous UFH, with an overall risk of HIT of about 0.2% in all heparin-exposed patients.[21]

Other studies in the literature quote frequencies as high as 1-5%.[22, 23, 24] High frequencies of HIT are especially common in surgical patients receiving prolonged postoperative thromboprophylaxis (eg, for 10-14 days following orthopedic surgery[25] or after coronary artery bypass and/or valve replacement surgery[21] ).

Mortality/Morbidity

HIT is a severe prothrombotic condition, with affected individuals having a greater than 50% risk of developing new thromboembolic events.[26] The mortality rate is approximately 20%, and approximately 10% of patients require amputations or suffer other major morbidity.[14, 27, 28]

A consecutive study with 108 hospitalized patients diagnosed with HIT showed that thrombotic complications occurred in about 29%. Early, severe falls in platelet counts in elderly patients receiving heparin appear to be associated with the development of thrombotic complications.[27]

Thrombosis associated with HIT can involve the arterial system, the venous system, or both. Thrombotic complications may include deep venous thrombosis, stroke, myocardial infarction, limb ischemia, and, rarely, ischemia of other organs. The thrombotic complications are fatal in about 29% of patients, and an additional 21% have to undergo limb amputations.[29]

Although HIT is a hypercoagulable disorder, patients remain at risk for major bleeding. A review by Pishko et al found that over a third of patients with HIT who were exposed to an alternative anticoagulant experienced a major bleeding event. Factors associated with increased risk of major bleeding included intensive care unit admission, platelet count <  25 × 109 /L, and renal dysfunction.[30]

Race

A study by Lewis et al reported that white patients had significantly less risk than nonwhite patients for thrombotic events, irrespective of the heparin-induced thrombocytopenia (HIT) presentation.[31] This study showed that nonwhites are approximately 2-3 times more likely than whites to progress to a HIT-associated thrombotic outcome, particularly with regard to new thromboses.[31]

Sex- and age-related differences in risk

Men have significantly less risk than women for thrombotic manifestations in HIT. Women diagnosed with HIT and thrombosis are 1.7 times more likely than men to have a new HIT-associated thrombotic event.[31]

The higher frequency of HIT in females was found most strikingly in patients treated with UHF. There was no relationship between sex and the risk for HIT in patients treated with low molecular weight heparin (LMWH).[32] LMWH in prevention of HIT may have the greatest absolute benefit in females undergoing surgical thromboprophylaxis.[32]

The mean age of patients with HIT is 62.3 ± 6.6 years.[33] In a retrospective study of 408 patients diagnosed with HIT, 66% of patients were older than 60 years.[34]  Obeng and colleagues, in a retrospective cohort study of 155 patients under 21 years old with sufficient data for 4Ts scoring for HIT, found that the prevalence of HIT was significantly lower in these patients than in adults.[35]

Until recently, the mortality rate in heparin-induced thrombocytopenia (HIT) has been reported as high as 20%, and a similar percentage of patients survived with major complications, including limb loss or stroke. Recent improvements in early diagnosis and treatment have resulted in a better prognosis, but rates of mortality and major complications of HIT are still as high as 6-10%.[36]

Complications

Possible complications of HIT include the following:

• Deep venous thrombosis
• Pulmonary embolism
• Myocardial infarction
• Occlusion of limb arteries (possibly resulting in amputation)
• Transient ischemic attack and stroke
• Skin necrosis
• End-organ damage (eg, adrenal, bowel, spleen, gallbladder, or hepatic infarction; renal failure)
• Death

Factors that increase the risk for heparin-induced thrombocytopenia (HIT) are as follows:

• Use of unfractionated heparin (UFH) rather than low molecular weight heparin (LMWH)

• Intravenous rather than subcutaneous administration of heparin

• Longer duration of heparin use

• Surgical (especially cardiac or orthopedic surgery) rather than medical patient

• Female sex

Clinical features that help distinguish HIT from other forms of thrombocytopenia include the timing of onset and the presence of thrombosis (eg, venous thromboembolism) or other sequelae. In contrast to other drug-induced immune thrombocytopenia syndromes, HIT is generally not associated with bleeding.[37]

Onset of HIT usually occurs 5-14 days after the start of heparin therapy. However, in patients with recent prior heparin exposure (within the past 100 days), persistence of circulating HIT antibodies may result in rapid-onset HIT, in which the platelet count falls within 24 hours of starting heparin.[2]

Onset of HIT may also occur after heparin cessation. Delayed-onset HIT should be considered when a patient presents with thrombosis and unexplained thrombocytopenia up to 3 weeks after the end of heparin therapy.[38]

Patients with heparin-induced thrombocytopenia (HIT) may present with unusual characteristic sequelae, including the following:

• Venous limb gangrene

• Bilateral adrenal hemorrhagic infarction

• Skin lesions at injection sites

• Acute systemic reactions following an intravenous heparin bolus

Acute systemic reactions include chills, fever, dyspnea, chest pain, and flushing.[39] Patients with HIT can also present with deep venous thrombosis (DVT) or pulmonary embolism. In patients with DVT who are started on warfarin, severe protein C depletion can develop, leading to warfarin-induced venous limb gangrene and likely loss of the limb.[40] Less often, patients may present with arterial thromboembolism (eg, stroke, myocardial infarction).

Heparin-induced thrombocytopenia (HIT) has three characteristic features that can distinguish it from other causes of thrombocytopenia.[37] First is the timing of the onset of thrombocytopenia; in most patients with HIT, the platelet count decrease begins from days 5 to 14 of heparin treatment. Second, the severity of the thrombocytopenia is usually mild to moderate, with platelet counts only rarely less than 15 × 109/L.

Third is the occurrence of large-vessel venous or arterial thrombosis in association with thrombocytopenia. Thrombosis precedes thrombocytopenia in up to 25% of patients with HIT.[47]

All patients should have baseline platelet counts measured before heparin treatment is started. Guidelines from the American College of Chest Physicians (ACCP) suggest basing further platelet counts on the patient’s risk of HIT. For patients whose risk is considered to be greater than 1% (eg, patients receiving unfractionated heparin after cardiac or orthopedic surgery), the ACCP suggests monitoring the platelet count every 2 or 3 days from days 4-14 of heparin therapy (or until heparin is stopped, whichever comes first).[47]

The ACCP suggests that platelet counts need not be monitored in patients receiving heparin whose risk of HIT is considered to be less than 1%. These include medical patients receiving low molecular weight heparin.

If the platelet count falls by over 50% of the baseline count, even if the nadir remains above 150 × 109/L, and/or a thrombotic event occurs, diagnostic tests should be performed.[47] Diagnostic tests for HIT consist of immunoassays and functional assays.

Immunoassays identify antibodies against heparin/platelet factor 4 (PF4) complexes. Functional assays measure the platelet-activating capacity of PF4/heparin-antibody complexes. Functional assays have greater specificity than immunoassays but are time-consuming and not widely available; many institutions offer only immunoassays.[44]

Immunoassays include solid-phase immunoassays (heparin/PF4 enzyme-linked immunosorbent assay [ELISA]), and particle gel immunoassays. Immunoassays are widely available and have a rapid turnaround time and high sensitivity (> 99%). However, they have poor specificity (30%-70%) because they also detect nonpathogenic antibodies.[48]

The specificity of ELISA can be enhanced by taking into account the optical density of the result. Higher absolute optical density values correlate with a clinical diagnosis of HIT. Warkentin and colleagues reported that weaker optical density values (0.4 to < 1.00) indicated a 5% or lower probability of a strongly positive result on functional testing with the serotonin release assay (SRA), whereas an optical density value of 2.0 or more resulted in an approximately 90% probability of a strong SRA result.[49]

Baroletti and colleagues found that in patients with clinically suspected HIT, a 1-unit increase in optical density values was associated with an approximate doubling in the odds of thrombosis by 30 days (odds ratio, 1.9; 95% confidence interval, 1.5-2.6; P=.0001). The proportion of patients with pulmonary embolism increased with higher values.[50]

In a study of surgical intensive care unit (ICU) patients, Berry and colleagues found that only 19% of patients with an optical density value of 0.4 or higher had a positive SRA result. Use of an optical density threshold of 2.0 proved more predictive of HIT, with a true positive rate of 65%.[46] Another study of surgical ICU patients, by Barada and colleagues, also recommended an optical density threshold of 2.0 or more, to reduce overdiagnosis and overtreatment.[45]

One proposed procedure to improve the specificity of the ELISA involves the addition of excess heparin to the sample. With a positive ELISA, a decrease in the optical density by 50% or more after the addition of excess heparin confirms the presence of heparin-dependent antibodies.[51]

Functional assays include the heparin-induced platelet aggregation assay (HIPA) and the serotonin release assay (SRA). Most laboratories that perform functional testing use HIPA, which is highly specific but which is also reported to be less sensitive than SRA. The availability of SRA is largely restricted to centers where HIT is a focus of research.

The SRA is based on HIT antibodies causing platelets to aggregate and release serotonin. For the test, platelets of normal donors are radiolabeled with carbon 14 (14C)-serotonin and are washed; these platelets are then mixed with patient serum, along with low (therapeutic) and high heparin concentrations. The test is considered positive if the sample causes a greater than 20% serotonin release at a (therapeutic) dosage of 0.1 U/mL heparin.

The14C-SRA is considered the "gold standard" assay for the detection of heparin-dependent antibodies in heparin-induced thrombocytopenia (HIT).[52] The sensitivity of SRA has ranged from 69% to 94%, and its specificity of SRA may be as high as 100%.[53, 54, 55]

HIPA is a platelet-activation test in which the patient's serum is mixed with donor platelets in the presence of heparin. Aggregation of the donor platelets indicates the presence of antibodies to the heparin–PF4 complex. Chong and colleagues reported that the mean sensitivity of HIPA may vary from 39% to 81%, depending on the heparin concentration and the reactivity of the platelets used, while specificity ranged from 82-100%.[55]

HIPA proved to be more sensitive than PF4 ELISA for laboratory confirmation of HIT, in a study involving serum samples from 146 patients examined for HIT. However, neither the HIPA nor the PF4 ELISA predicted thrombotic risk.[56]

Silent deep venous thrombosis (DVT) is common in patients with HIT. Therefore, bilateral lower extremity compression ultrasonography may be considered in these patients, even in the absence of clinical evidence of lower-limb DVT.[57]

In patients with HIT who experience thrombosis, different vascular imaging studies can be used to document the thrombotic lesions, including ultrasonography and angiography (see the images below). Multislice computed tomography (MSCT) scanning has been utilized in some patients with multiple thrombosis.[58]

Ultrasonographic image of a deep vein thrombosis (DVT).

Ultrasonographic image of a deep vein thrombosis (DVT).

Sequential images demonstrate treatment of iliofemoral deep venous thrombosis due to May-Thurner (Cockett) syndrome. Far left: View of the entire pelvis demonstrates iliac occlusion. Middle left: After 12 hours of catheter-directed thrombolysis, an obstruction at the left common iliac vein is evident. Middle right: After 24 hours of thrombolysis, a bandlike obstruction is seen; this is the impression made by the overlying right common iliac artery. Far left: After stent placement, image shows wide patency and rapid flow through the previously obstructed region. Note that the patient is in the prone position in all views. (Right and left are reversed.)

Sequential images demonstrate treatment of iliofemoral deep venous thrombosis due to May-Thurner (Cockett) syndrome. Far left: View of the entire pelvis demonstrates iliac occlusion. Middle left: After 12 hours of catheter-directed thrombolysis, an obstruction at the left common iliac vein is evident. Middle right: After 24 hours of thrombolysis, a bandlike obstruction is seen; this is the impression made by the overlying right common iliac artery. Far left: After stent placement, image shows wide patency and rapid flow through the previously obstructed region. Note that the patient is in the prone position in all views. (Right and left are reversed.)

Ventilation-perfusion scan. Left image: Posterior view of normal findings on ventilation-perfusion scan. Right image: Posterior view of a perfusion scan that reveals a perfusion defect in the left upper quadrant. The defect in the middle of the image is due to the position of the heart.

Ventilation-perfusion scan. Left image: Posterior view of normal findings on ventilation-perfusion scan. Right image: Posterior view of a perfusion scan that reveals a perfusion defect in the left upper quadrant. The defect in the middle of the image is due to the position of the heart.

Helical computed tomography scan of the pulmonary arteries. A filling defect in the right pulmonary artery is present, consistent with a pulmonary embolism.

Helical computed tomography scan of the pulmonary arteries. A filling defect in the right pulmonary artery is present, consistent with a pulmonary embolism.

If heparin-induced thrombocytopenia (HIT) is suspected, the first step is to discontinue all heparin products immediately and avoid any further exposure.[42] An order to simply "discontinue heparin" may not necessarily prevent heparin exposure, because unfractionated heparin (UFH) is commonly used to flush catheters.

Warfarin may cause microthrombosis in patients with HIT. These patients typically present with an international normalized ratio (INR) greater than 4, which corresponds to severe protein C depletion. Initiation of warfarin should be postponed until substantial platelet recovery. Preferably, warfarin should not be started before the platelet count exceeds 150 × 109/L.[2, 59] If warfarin has already been started, vitamin K should be given.[47]

Platelet transfusions should be avoided in HIT, as they may increase the thrombogenic effect.[59] Guidelines from the American College of Chest Physicians (ACCP) suggest limiting platelet transfusions to patients with severe thrombocytopenia who are experiencing bleeding or undergoing an invasive procedure with a high risk of bleeding.[47]

Patients with HIT are at high risk for thrombotic events and should be treated with alternative anticoagulants, typically a direct thrombin inhibitor (DTI). The US Food and Drug Administration (FDA) has approved the DTI argatroban (Acova) for prophylaxis and treatment of thrombosis in patients with HIT, and as an anticoagulant in patients who are undergoing percutaneous coronary intervention (PCI) and have, or are at risk for, HIT or HIT with thrombosis (HITT). The DTI bivalirudin (Angiomax) is also approved for this indication in PCI (although it has only been studied in patients receiving concomitant aspirin) and it has orphan drug designation for use as an anticoagulant in patients with or at risk of HIT or HITT. The DTI lepirudin was discontinued by the manufacturer in 2012.

The indirect factor Xa inhibitor fondaparinux (Arixtra) is not approved for use in HIT, but some experts consider it an important treatment option, especially in stable, non–critically ill patients.[57, 60, 61]

Another indirect factor Xa inhibitor, danaparoid, is no longer available in the United States but is used in Canada, Europe, and Australia. The ACCP recommends the use of fondaparinux in pregnant patients with HIT if danaparoid is not available, as this agent does not cross the placenta, but otherwise assigns a low evidence rating to treatment of HIT with fondaparinux.[47]

Several novel oral anticoagulants exist (eg, rivaroxaban, dabigatran, apixaban), and preliminary evidence suggests that they may be beneficial for HIT, particularly in cases refractory to standard therapies.[62, 63, 64, 65] However, these agents have not been fully assessed for treatment of patients with HIT and none have FDA approval for use in HIT.[57, 66]

High-dose intravenous immunoglobulin (IVIG) has been used as an adjunct to anticoagulation in patients with HIT. At a usual dose of 1 g/kg for 2 days, IVIG is typically effective in interrupting platelet activation by HIT antibodies and results in a rapid increase in the platelet count. IVIG has also been used to prevent acute HIT when re-exposure to heparin is planned in antibody-positive patients.[67]

For patients with heparin-induced thrombocytopenia (HIT) who have only isolated thrombocytopenia, therapeutic doses of alternative anticoagulants should be continued until the platelet counts recover to a stable plateau. Because the risk of thrombosis remains high for 2-4 weeks after treatment is initiated, consideration should be given to continuing anticoagulant therapy with an alternative agent or warfarin for up to 4 weeks.

For HIT patients with thrombosis, therapy with an alternative anticoagulant should be followed by a transition to warfarin, but only after platelet counts have recovered to above 150 x 109/L. Oral anticoagulants should be initiated at low doses (eg, ≤ 5 mg/day of warfarin), and an overlap with a direct thrombin inhibitor for at least 5 days should be planned until the international normalized ratio (INR) has been in the therapeutic range for at least 48 hours. Anticoagulation is typically continued for 3 months.[57]

For patients with a past history of HIT who no longer have circulating HIT antibodies, the American College of Chest Physicians suggests that short-term (intraoperative only) heparin can be used for cardiac surgery, but recommends bivalirudin or argatroban for cardiac catheterization or percutaneous coronary intervention. Patients with persistent HIT antibodies who require cardiac surgery should not receive heparin.[57, 47]

Treatment of heparin-induced thrombocytopenia (HIT) involves discontinuing heparin and starting a non-heparin anticoagulant. American Society of Hematology guidelines suggest that some agents may be preferred in certain circumstances, as follows[57] :

• Patients with critical illness, increased bleeding risk, or increased potential need for an urgent procedure - Argatroban or bivalirudin (because of their shorter duration of effect)
• Clinically stable patients - Fondaparinux or a direct oral anticoagulant (DOAC) (because of their ease of administration, lack of need for lab monitoring, and feasibility of outpatient use) 
• Life-threatening or limb-threatening thrombosis - Argatroban, bivalirudin, danaparoid, or fondaparinux
• Moderate or severe hepatic dysfunction (Child-Pugh Class B and C) - Avoid argatroban or use a reduced dose; avoid DOACs

The choice of non-heparin anticoagulant may also be influenced by drug factors (eg, availability, cost, route of administration), patient factors (eg, kidney function, liver function), and clinician experience.[47, 57]  For example, in patients with HIT who have thrombosis and renal insufficiency, American College of Chest Physicians (ACCP) guidelines suggest using argatroban; in those with normal kidney function, the ACCP suggests using argatroban or danaparoid.[47]

Argatroban (Acova)

Argatroban is a DTI; it inhibits fibrin formation, platelet aggregation, and activation of coagulation factors V, VIII, XIII, and protein C. The dose is 2 mcg/kg/min, adjusted by the activated partial thromboplastin time (aPTT) with a target of 1.5-3 times the baseline.

Argatroban is the ideal alternative to heparin for patients receiving dialysis, because it is not excreted by the kidneys and does not require dose adjustment in those patients.[51] However, because argatroban is processed by the liver, the initial dose should be reduced by 75% in patients with liver dysfunction.

Bivalirudin (Angiomax, Angiox)

Bivalirudin is a competitive, direct inhibitor of thrombin that inhibits both free and clot-bound thrombin and thrombin-induced platelet aggregation. This agent is approved for use in patients who are undergoing percutaneous coronary intervention (PCI) and have, or are at risk for, HIT or HIT with thrombosis (HITT).

Fondaparinux (Arixtra)

Fondaparinux is a synthetic anticoagulant that works by inhibiting factor Xa, a key component involved in blood clotting. It provides a highly predictable response. Bioavailability is 100%, has a rapid onset of action, and a half-life of 14-16 h, allowing for sustained antithrombotic activity over 24-h period. Fondaparinux does not affect prothrombin time or aPTT, nor does it affect platelet function or aggregation.

Warfarin (Coumadin, Jantoven)

Warfarin interferes with hepatic synthesis of vitamin K–dependent coagulation factors. It is used for prophylaxis and treatment of venous thrombosis, pulmonary embolism, and thromboembolic disorders. To avoid warfarin-induced venous limb gangrene in patients with HIT and deep venous thrombosis, warfarin should not be started until the platelet count has risen above 150 x 109/L.[1, 49] Tailor the dose of warfarin to maintain an International Normalized Ratio (INR) in the range of 2 to 3.