Erratum

In CATH, Volume 19, Issue number 2, March/April 2013, Chapter 19 titled “Heparin-Induced Thrombocytopenia” was inadvertently omitted.

The omitted chapter is published in the current issue, CATH 19:3 May/June. For the complete updated CATH 19:2 March/April issue, please log on to http://cat.sagepub.com.

SAGE deeply regrets the error.

Heparin-Induced Thrombocytopenia

General Considerations

Heparin-induced thrombocytopenia (HIT) is an important adverse effect of heparin. HIT is a life-threatening prothrombotic, immune-mediated coagulopathy caused by antibodies that bind to the complex of platelet factor 4 (PF4) and heparin. ¹ HIT occurs most frequently after cardiac or orthopedic surgery or in medical patients^{2 –10} but can also be found in other patient populations and clinical settings.^{11 –15} Progression to overt thrombosis, which can occur anywhere throughout the venous and arterial circulation, is the most serious complication of HIT, as it often leads to amputation or death.^{16 –19} Spontaneous bleeding and petechiae are rare.

HIT (also known as HIT type II) needs to be distinguished from other more common and benign causes of thrombocytopenia, such as HIT type I and pseudothrombocytopenia. HIT type I is a transient but self-limited fall in platelet count that occurs in up to 30% of the treated patients. It results from a nonimmunological mechanism in the first 24 hours of receiving heparin and resolves within 24 to 48 hours. ²⁰

The frequency of HIT is influenced by several factors. The risk of developing HIT is higher when exposed to unfractionated heparin (UFH; bovine more than porcine ²¹ ) than low-molecular-weight heparin (LMWH)^{7,22 –27} and is more duration dependent than dose dependent.^8,28,29 However, HIT can occur with a higher frequency in LMWH-treated patients who were previously exposed to UFH. ²⁹ HIT due to LMWH is as severe as UFH-induced HIT. ²² HIT can also occur with prophylactic doses of heparin ²⁸ and heparin from exogenous sources (eg, heparin flushes and heparin-coated catheters). ³⁰ Preventive measures include the use of LMWH, fondaparinux, and nonheparin anticoagulants rather than UFH for postsurgical prophylaxis, use of porcine rather than bovine UFH, and avoiding unnecessary and prolonged exposure to UFH.

The diagnosis of HIT is based on the clinical findings and platelet count. In patients being treated or having been recently treated with heparin, HIT should be suspected on the basis of a 30% decrease in platelet count from baseline in the absence of other reasons for thrombocytopenia.^1,31,32 The diagnosis can be made if the platelet count reduction is 50% of the baseline, assuming no other reasons for thrombocytopenia.^1,31,32 An abrupt decrease in platelet count in the absence of other causes, which does not result in thrombocytopenia (eg, platelet count may fall from 350 to 175 × 10⁹/L) and unexplained thrombosis are also characteristics of HIT.^1,31,32 Symptoms typically appear 4 to 14 days after exposure to UFH^33,34 or 8 to 14 days after exposure to LMWH. ²² Patients who received heparin within the prior 100 days can have an immediate rapid onset HIT when restarting UFH or LMWH.^33,34 Delayed onset HIT has been observed with symptoms appearing several days after discontinuation of UFH.^1,35

The diagnosis of HIT is difficult in patients who underwent surgery, as postoperative thrombocytopenia is frequently present after a surgical procedure. It is particularly difficult after cardiac surgery, because the platelet count always falls following cardiac surgery using cardiopulmonary bypass. In these patients, HIT should be suspected if the platelet count recovery in the immediate postoperative period is interrupted by a sudden and marked platelet count decrease in 5 to 10 days postoperation (a biphasic platelet count pattern).^36–38 However, HIT cannot be definitely excluded in these patients if there is a monophasic pattern of persistent postoperative thrombocytopenia.^36–38

Another clinical presentation of HIT that can be challenging is where a patient has only mild thrombocytopenia receiving heparin or LMWH treatment. Such patients need to be individually assessed for their risk of having HIT considering past exposure to heparin, competing causes for thrombocytopenia, and new thrombosis. The level of risk will determine whether or not to continue heparin or LMWH treatment, while laboratory testing is sent to confirm the diagnosis. ³⁹

Clinical scoring systems are available and continue to be developed to assist in the diagnosis of HIT.^36,39–41

A clinical diagnosis of HIT should be confirmed by a laboratory assay that detects heparin-dependent antibodies. Pathologic HIT immune complexes are composed of the PF4-heparin complex bound to an immunoglobulin (Ig) G.^{42 –46} These complexes bind to platelet FcγIIa receptors (CD 32) inducing platelet activation, aggregation, and generation of platelet microparticles.^47,48 IgA and IgM have also been identified in patients with HIT. ⁴⁹ The HIT antibodies provoke leukocyte and endothelial cell activation that augment both the hypercoagulable and the inflammatory states.^{44,50 –54} This combined cellular activation leads to a burst of thrombin generation. ⁵⁵ In all patients at risk of thrombosis, those with HIT are at highest risk (>30%). ³¹ Nondrug factors, such as type of surgery, severity of trauma, severity of thrombocytopenia (particularly at baseline), renal impairment, low cardiac output, and timing of first anticoagulant dose, also influence the risk of developing HIT and related clinical outcomes.^{56 –59} The association of HIT antibodies, in the absence of thrombocytopenia and thrombosis, with future cardiovascular and other thrombotic events has been reported and remains under investigation. ⁶⁰

There are 2 types of laboratory assays that detect heparin-dependent antibodies. These are platelet function tests (serotonin release and platelet aggregation assays) and immunoassays that detect antibodies to the PF4-heparin complex.^{61 –64} Each test has particular performance characteristics and provides unique information, so that appropriate use and knowledgeable interpretation of the test results are important.^{64 –67} Platelet function assays that use washed platelets have a better sensitivity than plasma-based assays, but false negative results can still be obtained. Immunoassays have a high rate of positive results that are not always associated with clinical HIT in the patient.^65,68–70 For immunoassays, the option to report the titer results rather than a simple positive or negative result and to utilize the high heparin concentration confirmatory step are gaining favor, as these provide a closer correlation with the risk of thrombosis and mortality in patients with HIT.^{26,71 –75} Exclusive reliance on laboratory tests for the diagnosis of HIT can lead to erroneous diagnostic conclusions.

Clinical trials and clinical experience have shown the direct thrombin inhibitors (DTIs) argatroban^{76 –82} and lepirudin^83,84 to be safe and effective in reducing the risk of thrombosis and associated morbidity or mortality in patients with HIT. These drugs do not cross-react with HIT antibodies. Development of antibodies to lepirudin has been observed in approximately 50% of the patients after 10 days of treatment, including severe anaphylactic reactions with fatal outcomes in cases of reexposure to lepirudin.^85,86 Dose adjustments for argatroban in specific populations^79,87,88 and for lepirudin in general^89–91 have been recently recommended. The DTI desirudin, which has the advantage of subcutaneous dosing, has been successfully used in a limited number of patients with HIT.^92,93 The DTI bivalirudin, which has a short half-life and enzymatic degradation, has been used for anticoagulation in patients with HIT during cardiac surgery.^94–96 The DTIs have also been used successfully in patients with HIT requiring invasive cardiac procedures.^97,98 The DTIs should be treated as individual drugs, as each has its own pharmacologic characteristics.

The heparinoid danaparoid has been successfully used to treat patients with HIT,^99–101 but there are reports that danaparoid cross-reacts with some HIT antibodies leading to treatment failures.^{101 –105} The synthetic heparin pentasaccharide, fondaparinux, has been shown to be useful for the management of patients with HIT through several small published case series and is gaining favor.^{106 –109}

The LMWHs can cross-react with most HIT antibodies and are contraindicated for use in patients with HIT.^27,110,111

Vitamin K antagonists (VKAs) are recommended for long-term treatment of HIT-associated thrombosis. ³¹ The VKAs are not recommended for use in the acute phase of HIT due to their potential to intensify the prothrombotic state from a transient protein C deficiency.^112,113 The VKAs should be initiated when platelet counts have normalized to a steady state, then brought on underbridged with a DTI.^114–116

There is emerging evidence that the newly developed small molecule anticoagulants including apixaban, dabigatran, edoxaban, otamixaban, and rivaroxaban may become new immediate and long-term treatment options for thrombosis in patients with HIT. ¹⁰⁹

Recommendations

Early diagnosis and treatment are important to improve clinical outcomes. Diagnosis of HIT is based on a comprehensive interpretation of clinical and laboratory information.

For the first 14 days of treatment, platelet counts should be performed every 2 to 3 days in patients treated with LMWH and daily if treated with UFH and if the patient’s risk of developing HIT is high (level of evidence: moderate). For medical and obstetric patients treated with LMWH exclusively and no prior exposure to UFH, it is no longer considered necessary to monitor the platelet count. Patients with comorbidities are at higher risk of poorer clinical outcomes. All clinical settings including the emergency department need to be aware of a patient’s history of HIT and prior UFH or LMWH exposure.

Several clinical scoring systems are available, which can help diagnose HIT. Laboratory testing should be performed when there is a strong suspicion of HIT (level of evidence: moderate). Laboratory tests are used to confirm a diagnosis of HIT, but negative results do not exclude the diagnosis. It is useful to perform a combination of tests and to repeat testing over a period of several days. Initial therapeutic decisions should not be dependent upon a positive laboratory test but should be based upon clinical findings, particularly thrombocytopenia and/or new thromboembolic events.

UFH and LMWH should be stopped when the diagnosis of HIT is strongly suspected or confirmed (level of evidence: high). It is not sufficient to merely remove the heparin. Due to the strong hypercoagulable state and high risk of thrombosis associated with HIT, it is recommended that all patients with HIT be treated with a nonheparin anticoagulant such as argatroban, lepirudin, or danaparoid (level of evidence: moderate). Differences between these drugs need to be considered when making a clinical treatment decision (eg, patient’s renal or liver clearance, drug pharmacokinetics, patient’s risk of bleeding, prior exposure of patient to lepirudin, physician’s experience with the drug, drug availability, acute anticoagulant need, long-term treatment, cross-reactivity of drug to HIT antibodies, etc). With danaparoid treatment, if daily platelet counts do not show signs of recovery within 3 days, it is mandatory to check for immune cross-reactivity of patient’s antibodies to danaparoid using a functional platelet assay and discontinue treatment if positive. Fondaparinux may be considered as a second-line agent in the management of patients with suspected HIT (level of evidence: low). The LMWHs are contraindicated in patients with HIT (level of evidence: moderate).

For long-term anticoagulation, a VKA can be used. To avoid warfarin-induced limb gangrene or skin necrosis in patients with HIT, the VKA should only be administered after rise in platelet counts, with substantial recovery to >100 × 10⁹/L or to pre-HIT values (level of evidence: low). Starting doses need to be low (5 mg warfarin and 6 mg phenprocoumon) and given with overlapping administration of argatroban, lepirudin, or danaparoid for at least 5 days.

For patients with HIT undergoing coronary artery interventional procedures, bivalirudin or argatroban anticoagulation is recommended (level of evidence: moderate). For special populations of patients with HIT requiring anticoagulation such as pregnant or pediatric patients or patients undergoing cardiac surgery or hemodialysis, specific drug and dose issues need to be considered. For postoperative cardiac surgery patients, LMWH, and not, UFH should be used to decrease the risk of developing HIT, and if HIT is suspected, a nonheparin anticoagulant should be used.