Complications of Heparin Anticoagulation Therapy

Heparin-Induced Thrombocytopenia

HIT is an uncommon but serious complication of heparin therapy. It is characterized by thrombocytopenia, platelet activation, and thrombosis. Two distinct types of HIT are described: type I and type II.

HIT type I (nonimmune HIT) is more common and more transient. It occurs in 10% to 30% of patients exposed to heparin and develops within a few days after heparin administration. There is minimal risk of thrombosis, and recovery occurs without any treatment. The pathogenesis is attributed to the platelet proaggregating effects of heparin, with a mild decrease in platelet count up to 30% of basal value.

HIT type II (immune-mediated HIT) is a true immunologic phenomenon, uncommon but dangerous, that occurs in 0.3% to 5% of patients after heparin exposure. It is associated with a high risk (20%–50%) of developing venous and arterial thrombotic events. Clinically, this results in HIT with thrombosis (HITT), a devastating condition that carries a mortality rate of up to 20%. Its onset from 5 to 14 days following the initiation of heparin therapy is a classic feature of HITT, with a notable exception with a more-rapid onset in patients who have received heparin in the previous 3 months. The latter phenomenon is caused by the presence of preexisting antibodies. HITT rarely occurs after day 15. The platelet count usually falls by more than 50% of basal value, with a median nadir of 55 × 10 ⁹ /L.

Despite thrombocytopenia, hemorrhagic events are rare. Thrombosis is the hallmark complication of this entity, including development of thrombi in arteries and veins, erythematous or necrotic skin at the site of heparin injection, or disseminated intravascular coagulation (DIC). The onset of HITT is independent of dosage, schedule, and route of administration of heparin. Unfractionated heparin (UFH) is associated with a higher incidence of HIT than low-molecular-weight heparin (LMWH).

A clinical scoring system (4 Ts) proposed by Warkentin and Heddle in 2003 can be used to assess the pretest probability of HIT type II based on four characteristic features: the degree of thrombocytopenia, the timing of the onset, the presence of new or progressive thrombosis, and whether an alternative cause of thrombocytopenia is likely ( Table 1 ). A moderate to high pretest probability mandates cessation of heparin and administering an alternative anticoagulant. With a low pretest probability, other reasons for thrombocytopenia or thrombosis should be sought.

The pathogenesis of HIT type II is based on immune-mediated antibody formation. The negatively charged heparin has high affinity to positively charged platelet factor 4 (PF4), found on the cell surface of platelets and endothelial cells. The heparin–PF4 complex serves as an immunogen generating HIT antibodies, most commonly IgG. The heparin–PF4–IgG complex activates platelets by binding to Fc receptors, causing release of prothrombotic platelet-derived microparticles, platelet consumption, and thrombocytopenia. However, only a subset of heparin–PF4 antibodies activate platelets, causing a clinical manifestation of HIT.

The diagnosis of HIT is primarily a clinical one, and treatment should not be delayed for patients with a high index of suspicion while awaiting laboratory tests for confirmation. However, common causes of iatrogenic thrombocytopenia should be kept in mind, including sepsis and medications such as antibiotics and H ₂ blockers. Heart surgery, cardiac devices such as intraaortic balloon pumps and left ventricular assistive devices, and some heart medications are also established causes of thrombocytopenia.

Two assays are available for detecting HIT antibodies. An enzyme-linked immunosorbent assay (ELISA) detects the presence of IgG, IgM, and IgA globulin classes. Only IgG-type antibodies are found to cause clinically relevant HIT. This assay is routinely used as a screening test because it is easy to perform, has a rapid turnover time, and has a high sensitivity (99%–100%). A negative result virtually rules out HIT. However, it lacks specificity owing to its inability to differentiate clinically relevant antibodies from nonspecific heparin–PF4 antibody complexes. Therefore, a positive test does not mean that HIT is present.

Functional assays, or platelet-activation assays, measure platelet activation against the heparin–PF4 antibody. The commonly available tests include platelet aggregation (PAT), heparin-induced platelet activation (HIPA), and serotonin-release assay (SRA). Among these, SRA is considered the gold standard for confirming HIT. It measures ¹⁴ C-serotonin release from donor platelets in response to the patient’s plasma as a measure of platelet activity. This method detects functional platelets with a high specificity (89%–100%).

Management of a patient with a history of HIT depends on the type of surgery and the indication for anticoagulation. Wherever possible, an alternative anticoagulant should be considered. However, because levels of heparin–PF4 HIT antibodies decline after 3 months, it is acceptable to administer heparin in a selected group of patients who had an episode of HIT more than 3 months previously. In this subset of patients, it is prudent to administer heparin as a periprocedural bolus rather than as an infusion. In patients with acute or active HIT or those with a history of HIT less than 3 months and/or lingering HIT antibodies, surgery should be delayed until HIT is resolved and antibodies are undetectable. In urgent situations, alternative anticoagulants should preferably be used intraoperatively.

Because the risk of thrombotic complications is high, treatment for suspected HIT should not be delayed. Heparin must be stopped immediately, including prophylactic or therapeutic subcutaneous UFH or LMWH, intravenous heparin catheter flushes, and heparin-coated indwelling central venous catheters or peripherally inserted central catheters. HIT should be considered in patients who are exposed to heparin who develop an abrupt decrease (>50%) in platelet count within 5 to 14 days of exposure and those who develop a new thrombotic event such as deep vein thrombosis, pulmonary embolism, stroke, myocardial infarction, limb thrombosis, or occlusion of coronary or extremity bypass grafts. HIT should be especially high on the differential diagnosis for patients who come to the hospital with unusual clots such as mesenteric thrombosis, cerebral sinus thrombosis, and bilateral hemorrhagic adrenal necrosis. A high index of suspicion must be maintained in patients with heparin-bonded polytetrafluoroethylene (PTFE) bypass grafts; although rare, HIT has been reported in such patients. Alternative anticoagulants should be administered promptly to prevent thromboembolic events. Several options are available.

Direct thrombin inhibitors (DTI) are the most commonly used medications for the treatment of HIT. Argatroban (GlaxoSmithKline, Research Triangle Park, NC) is indicated for the treatment and prophylaxis of HIT-associated thrombosis. It is a synthetic DTI that reversibly binds to the thrombin-active site, inhibiting thrombin-catalyzed or thrombin-induced reactions and platelet aggregation. It has a half-life of 45 minutes and is metabolized in the liver. The 2012 American College of Chest Physicians (ACCP) guidelines suggest ommision of an initial bolus, and an initial infusion rate of ≤2 mcg/kg/min. For patients who have heart failure, multiple organ system failure, or severe anasarca, or who are post cardiac surgery, the initial infusion rate is recommended at between 0.5 and 1.2 mcg/kg/min, with subsequent q2h adjustments using the aPTT (target aPTT 1.5-3 times ptient baseline.

Lepirudin (Refludan; Bayer Healthcare, West Haven, CT), also a DTI, is a recombinant hirudin analogue approved for anticoagulation treatment of patients with HIT. Lepirudin irreversibly binds and inactivates thrombin, without a direct effect on platelets. It is renally excreted and has a half-life of 1 hour. Current dosing recommendations are 0.2 mg/kg as a bolus followed by ≤0.10 mg/kg/hour continuous infusion rate, with a lower infusion rate for patients with renal failure, to achieve a target therapeutic dose range of aPTT at 1.5 to 2.0 times the baseline value. A major drawback of lepirudin is its immunogenicity. Approximately 40% of HIT patients treated with lepirudin develop anti-hirudin antibodies that can cause a fatal anaphylactic reaction.

Yet another DTI, bivalirudin (Angiomax; The Medicines Company, Parsippany, NJ), is approved for anticoagulation during percutaneous coronary intervention (PCI) in patients with confirmed or suspected HIT. It is also a recombinant hirudin analogue that reversibly binds and inactivates thrombin. Bivalirudin is renally excreted and has a short half-life of 25 minutes. Current ACCP guidelines suggest no inital bolus and a starting infusion rate of 0.15-0.20 mg/kg/h, with a target of 1.5-2.5 times patient's baseline aPTT. It has been reported that antibodies against lepirudin can cross react with bivalirudin, and thus caution is recommended for its use in patients previously treated with lepirudin.

Fondaparinux (Arixtra, GlaxoSmithKline, Brentford, Middlesex, UK), a synthetic pentasaccharide, has been recognized as a new option to treat HIT. It selectively inhibits factor Xa by binding with high affinity to antithrombin III. It is renally excreted and has a half-life of 17 to 20 hours. Its molecular structure is similar to heparin’s but with a smaller molecular weight, and it has no cross reactivity with HIT antibodies. Its pharmacokinetic properties are favorable, with complete subcutaneous absorption rapid and independent of dose, and it requires only once-a-day subcutaneous administration. Current dosing guidelines for patients who weigh <50 kg: 5 mg subcutaneous (SC) daily; for those who weigh 50-100 kg: 7.5 mg SC daily; and for those who weigh >100 kg: 10 mg SC daily.

Vitamin K antagonists, most commonly sodium warfarin, are often used to provide continued anticoagulation treatment when platelet counts have reached normal levels after initiation of deep thrombin inhibition. Warfarin causes a transient procoagulant state and can increase the risk of micro vascular thrombosis in HIT, causing venous limb gangrene and skin necrosis. Therefore, its administration should be delayed until thrombocytopenia is resolved (>150,000/μL). It should be introduced with a minimum overlap with DTI of at least 5 days, and DTI should be continued until the international normalized ratio (INR) has reached an intended target range. Argatroban increases the INR, and thus it should only be stopped when the INR on combined argatroban and warfarin is >4.

Placement of an inferior vena cava filter is not indicated in patients who develop DVT from HIT. Treatment includes cessation of all forms of heparin and immediate initiation of DTI.