Library › Pharmacology › Parenteral anticoagulants and HIT

Parenteral anticoagulants and heparin-induced thrombocytopenia

Unfractionated heparin, low-molecular-weight heparin, fondaparinux, bivalirudin, and argatroban from pharmacology and dosing through procedural use, and heparin-induced thrombocytopenia from mechanism through diagnosis, treatment, and cardiac procedures.

Classes and mechanisms of parenteral anticoagulants

Parenteral anticoagulants act on thrombin (factor IIa) and factor Xa through two mechanisms. The indirect agents (unfractionated heparin [UFH], low-molecular-weight heparin [LMWH], and fondaparinux) have no intrinsic anticoagulant activity: they bind antithrombin, a circulating serine protease inhibitor, and accelerate its inactivation of thrombin and factor Xa. The direct thrombin inhibitors (bivalirudin and argatroban) bind thrombin without a cofactor. Indirect agents depend on antithrombin and are weakened by its deficiency. They inhibit free thrombin but have limited access to fibrin-bound thrombin. The heparins induce the platelet factor 4 (PF4) antibodies that cause heparin-induced thrombocytopenia (HIT); fondaparinux does not induce them at a clinically important frequency. Direct thrombin inhibitors inhibit fibrin-bound thrombin, do not bind PF4, and are the agents designed for patients with HIT.

The heparins are polysaccharides of varying length. Only chains that contain a specific pentasaccharide sequence bind antithrombin, and only about one-third of UFH molecules carry it. Inhibition of factor Xa requires only the pentasaccharide to change the conformation of antithrombin. Inhibition of thrombin requires a chain of at least 18 saccharides, long enough to bind antithrombin and thrombin together in a ternary bridge. The ratio of anti-Xa to anti-IIa activity therefore rises as the chains shorten: it is 1:1 for UFH, about 2:1 to 4:1 for the LMWHs, and infinite for fondaparinux, which is the pentasaccharide alone.

Properties of the parenteral anticoagulants used in cardiovascular medicine, from the US prescribing information and the 2012 CHEST review of parenteral anticoagulants. aPTT, activated partial thromboplastin time; ACT, activated clotting time.

Six factors govern the choice of agent: the need for rapid reversal or titration, which favors UFH; kidney function, because LMWH, fondaparinux, and bivalirudin accumulate; liver function, because argatroban accumulates; HIT or a history of it, which calls for a direct thrombin inhibitor or fondaparinux; the procedural setting, in which fondaparinux alone is unsafe for PCI; and the availability of a reliable monitoring assay. Oral anticoagulants, including the transition from parenteral therapy in venous thromboembolism, are covered in the oral anticoagulant hub, and management of bleeding and reversal across all agents in the bleeding, reversal, and periprocedural hub.

Unfractionated heparin: structure, binding, and pharmacokinetics

Pharmaceutical heparin is extracted from porcine intestinal mucosa. Its chains range in molecular weight from about 3,000 to 30,000 daltons, with a mean near 15,000 (about 45 saccharide units). Besides inhibiting thrombin and factor Xa, the heparin-antithrombin complex inactivates factors IXa, XIa, and XIIa, and heparin releases tissue factor pathway inhibitor from the endothelium. Heparin also binds to platelet factor 4, von Willebrand factor, fibrinogen, factor VIII, acute-phase proteins, endothelial cells, and macrophages. This nonspecific binding is the source of almost every limitation of the drug: it lowers the fraction available to antithrombin, varies between patients and over time, and makes the dose-response unpredictable.

Clearance proceeds by two mechanisms. At low doses, a rapid saturable mechanism dominates, in which heparin binds endothelial cells and macrophages and is depolymerized. Once those sites are saturated, a slower nonsaturable renal mechanism takes over. The result is nonlinear kinetics: the apparent half-life lengthens with dose, from about 30 minutes after an intravenous bolus of 25 U/kg to about 60 minutes after 100 U/kg and about 150 minutes after 400 U/kg. Continuous infusion after a bolus reaches steady state quickly, and the anticoagulant effect disappears within a few hours of stopping the infusion. Therapeutic intravenous heparin needs no dose adjustment in kidney failure, which makes it the parenteral agent of choice when creatinine clearance is very low or unstable.

Subcutaneous heparin has lower and more variable bioavailability than intravenous heparin. Prophylactic doses of 5,000 units every 8 or 12 hours prevent venous thromboembolism in medical and surgical patients. Higher fixed subcutaneous doses can provide therapeutic anticoagulation: in a randomized trial of 708 patients with acute venous thromboembolism, subcutaneous UFH given as 333 U/kg initially and 250 U/kg every 12 hours without monitoring was as effective (recurrence 3.8% versus 3.4%) and as safe (major bleeding 1.1% versus 1.4%) as LMWH, and most patients in both groups were treated entirely at home.

Weight-based heparin dosing and adjustment nomograms

Weight-based dosing reaches the therapeutic range faster than fixed dosing. In the trial that introduced the weight-based nomogram, 115 patients receiving intravenous heparin for venous or arterial thromboembolism or unstable angina were randomized to a bolus of 80 U/kg with an infusion of 18 U/kg per hour, or to a fixed 5,000-unit bolus with 1,000 units per hour. The weight-based regimen achieved a therapeutic aPTT more rapidly and with fewer subtherapeutic values, and recurrent thromboembolism was less frequent. The aPTT was measured every 6 hours, approximately four half-lives after each change, and the dose was adjusted by the protocol in the table below.

Arterial indications use lower doses. Patients with acute coronary syndromes (ACS) receive aspirin and a P2Y12 inhibitor, often with fibrinolytic therapy or PCI, and high heparin doses in that setting increase bleeding without ischemic benefit. The 2025 ACC/AHA ACS guideline therefore specifies a capped regimen.

Unfractionated heparin regimens by indication. ACS doses from the 2025 ACC/AHA ACS guideline (Table 10); PCI doses and activated clotting time (ACT) targets from the 2011 ACCF/AHA/SCAI PCI guideline; venous thromboembolism regimens from the weight-based nomogram trial and the fixed-dose subcutaneous trial; the bypass ACT target from the International Society on Thrombosis and Haemostasis (ISTH) heparin resistance communication. ACT values depend on the device and activator.

Weight-based heparin nomogram for venous thromboembolism (starting dose 80 U/kg bolus, 18 U/kg/h), as published in 1993. The seconds refer to the reagent used in that trial; each laboratory converts the ratios to its own reagent-specific range.

Calibration of the therapeutic range. The aPTT response to a given heparin level differs widely among reagents and coagulometers, so a fixed range such as 1.5 to 2.5 times control can be subtherapeutic with one reagent and supratherapeutic with another. The 2012 CHEST review of parenteral anticoagulants states that the therapeutic aPTT range at each institution is adapted to the responsiveness of its reagent and coagulometer; in the study that established the range, an aPTT ratio of 1.5 to 2.5 corresponded to a heparin level of 0.3 to 0.7 IU/mL by anti-factor Xa assay. A nomogram is valid only with the range for which it was written.

Body weight and dose caps. Weight-based nomograms use actual body weight. Many institutions cap the initial bolus and infusion in severe obesity, because the volume of distribution of heparin rises less than proportionally with fat mass; practice varies, and anti-Xa monitoring is preferred when weight exceeds the range studied. The ACS regimen is capped by guideline at 4,000 units for the bolus and 1,000 units per hour for the infusion, which applies to any patient above about 67 kg for the bolus and 83 kg for the infusion.

Protocol-driven management. Nursing- or pharmacist-managed nomograms with fixed sampling intervals reach the therapeutic range faster and with fewer errors than physician-ordered dosing. Daily platelet counts are part of the protocol whenever the patient is at intermediate or high risk of HIT.

Doses from the 2025 ACC/AHA ACS guideline (Table 10), the weight-based heparin nomogram, and the US prescribing information. Creatinine clearance is estimated by the Cockcroft-Gault equation. Enoxaparin doses are rounded to the nearest milligram, and the administered dose is rounded to the nearest syringe graduation; the enoxaparin regimen for fibrinolysis caps the first two doses at 100 mg (75 mg at 75 years or older). The argatroban rate in mg/h assumes no loading dose. The bivalirudin infusion shown is the procedural rate; a full-dose infusion for 2 to 4 hours after primary PCI is part of the STEMI regimen.

Monitoring heparin: aPTT, anti-Xa activity, and activated clotting time

Activated partial thromboplastin time. The aPTT is a clotting time that is prolonged by heparin's inhibition of thrombin, factor Xa, and factor IXa. It is inexpensive and universally available but measures heparin effect indirectly and is affected by factors unrelated to heparin. Elevated factor VIII and fibrinogen in acute inflammation, pregnancy, and cancer shorten the aPTT and suggest heparin resistance when the heparin level is therapeutic. A lupus anticoagulant, factor XII deficiency, liver disease, warfarin, and consumptive coagulopathy prolong the baseline aPTT, so the aPTT cannot be used to monitor heparin in these settings. Samples drawn from a line through which heparin is infused give falsely long values.

Anti-factor Xa activity. The chromogenic anti-Xa assay measures the ability of the patient's plasma to inhibit a fixed amount of factor Xa, and the result is reported in IU/mL against a heparin-specific calibrator. It is not affected by factor VIII, fibrinogen, factor deficiencies, or the lupus anticoagulant. The therapeutic range for intravenous UFH is 0.3 to 0.7 IU/mL. Monitoring heparin infusions by anti-Xa activity requires fewer dose changes and produces fewer results outside the range. Interference arises from hyperbilirubinemia and hemolysis in some assays, and from any other anti-Xa drug in the circulation: a patient switched from apixaban or rivaroxaban to heparin can have falsely high heparin anti-Xa levels for a day or more, and the aPTT is used during that interval. Some assays add exogenous antithrombin, which overestimates heparin effect when the patient's own antithrombin is low.

Activated clotting time. The ACT is a whole-blood clotting time measured at the bedside with a contact activator. It responds linearly to the high heparin concentrations used in PCI and cardiopulmonary bypass, at which the aPTT becomes unmeasurable. ACT results differ between devices: the HemoTec device gives values about 50 seconds lower than the Hemochron device at the same heparin level, which is why procedural targets name the device. ACT is prolonged by hypothermia, hemodilution, thrombocytopenia, and glycoprotein IIb/IIIa inhibitors, and it is unreliable at low heparin levels.

Laboratory monitoring of parenteral anticoagulants. Peak LMWH anti-Xa activity is measured about 4 hours after a subcutaneous dose, after the third or fourth dose. DOAC, direct oral anticoagulant; ECMO, extracorporeal membrane oxygenation; GPI, glycoprotein IIb/IIIa inhibitor.

Heparin resistance and antithrombin deficiency

Heparin resistance describes the need for unusually large heparin doses to reach a target aPTT or ACT. For intravenous therapy, it has classically been defined as a requirement of 35,000 units or more in 24 hours. For cardiopulmonary bypass, a 2024 ISTH Scientific and Standardization Committee communication proposed a standard definition: failure to reach an ACT of 480 seconds or more after 500 U/kg of heparin.

The mechanisms fall into two groups. True pharmacologic resistance results from antithrombin deficiency (inherited, or acquired from consumption in disseminated intravascular coagulation, extracorporeal circuits, prolonged heparin therapy, nephrotic syndrome, liver failure, and L-asparaginase), from increased heparin clearance, and from high concentrations of heparin-binding proteins in acute illness. Apparent (pseudo) resistance occurs when the heparin level is therapeutic but high factor VIII and fibrinogen shorten the aPTT; it is common in critical illness, pregnancy, cancer, and the postoperative state.

A heparin anti-Xa level separates true from apparent resistance: a therapeutic anti-Xa level with a short aPTT identifies the apparent form. In a randomized trial of 131 patients with venous thromboembolism who required 35,000 units or more of heparin per day, monitoring by anti-Xa activity (target 0.35 to 0.67 U/mL) instead of aPTT used significantly less heparin. Recurrence (4.6% versus 6.1%) and bleeding (1.5% versus 6.1%) were similar. True resistance from antithrombin deficiency before cardiopulmonary bypass is treated with antithrombin concentrate (plasma-derived or recombinant) in preference to fresh frozen plasma, which is a less concentrated source. When resistance persists despite heparin escalation and antithrombin repletion, a direct thrombin inhibitor (bivalirudin or argatroban) provides anticoagulation that does not depend on antithrombin.

Adverse effects of heparin other than HIT

Bleeding. Bleeding is the principal toxicity of every anticoagulant. With heparin, the risk rises with the intensity of anticoagulation, concomitant antiplatelet or fibrinolytic therapy, recent surgery or trauma, older age, female sex, low body weight, and kidney disease. In ACS registries, excess dosing is a recognized cause of bleeding, which is the basis of the capped ACS regimen.

Osteoporosis. Long-term heparin, particularly UFH for more than 1 to 3 months, reduces bone formation and increases resorption. Symptomatic vertebral fractures have been reported in pregnant women receiving months of UFH. LMWH causes less bone loss, which is one reason it is preferred for prolonged therapy in pregnancy.

Hyperkalemia. Heparin suppresses adrenal aldosterone synthesis. Hyperkalemia appears within days, is dose related, and is most likely with diabetes, kidney disease, and concomitant renin-angiotensin system inhibitors, potassium-sparing diuretics, or potassium supplements. Prophylactic subcutaneous doses can produce it.

Other effects. Asymptomatic elevation of transaminases is common and resolves after heparin is stopped. Hypersensitivity reactions and injection-site hematomas occur. Erythematous plaques at injection sites are most often delayed-type hypersensitivity but sometimes the first sign of HIT. Alopecia is rare. Non-immune heparin-associated thrombocytopenia, formerly called type I HIT, is a mild fall in platelet count within the first 2 days that resolves without stopping heparin; it carries no thrombotic risk and is distinguished from immune HIT by its timing and magnitude.

Reversal with protamine. Protamine sulfate, a basic protein derived from fish sperm, forms an inactive salt with heparin. One milligram neutralizes about 100 units of heparin, the dose is based on the heparin given in the preceding 2 to 3 hours because of the short half-life, and no more than 50 mg is given over 10 minutes. Protamine neutralizes all anti-IIa activity of LMWH but at most about 60% of its anti-Xa activity, and it does not neutralize fondaparinux. Hypotension, bradycardia, pulmonary hypertension, and anaphylactoid reactions occur with rapid infusion and in patients with prior protamine exposure (including neutral protamine Hagedorn [NPH] insulin) or fish allergy. Dosing for bleeding, the reversal of other agents, and the timing of interruption around procedures are covered in the bleeding, reversal, and periprocedural hub.

Low-molecular-weight heparins: pharmacology

LMWHs are produced by chemical or enzymatic depolymerization of UFH into chains with a mean molecular weight of about 4,000 to 5,000 daltons. Fewer than half of their pentasaccharide-containing chains are long enough to bridge antithrombin to thrombin, so anti-Xa activity predominates. Enoxaparin is the LMWH used in US cardiovascular practice; dalteparin is used mainly for venous thromboembolism and cancer-associated thrombosis, and tinzaparin is no longer marketed in the United States.

Their shorter chains bind less to plasma proteins, endothelium, and macrophages. The consequences are near-complete subcutaneous bioavailability, a peak anti-Xa effect 3 to 5 hours after injection, dose-independent renal clearance, a longer half-life, and a predictable dose-response that permits weight-based dosing without routine monitoring. Less binding to platelets and PF4 explains the lower incidence of HIT. The disadvantages follow from the same properties: LMWHs accumulate in kidney disease, protamine reverses them only partly, their prolonged effect complicates urgent procedures and neuraxial anesthesia, and the aPTT does not measure it.

Enoxaparin dosing by indication, kidney function, body weight, and age

Enoxaparin dosing from the US prescribing information and the 2025 ACC/AHA ACS guideline (Table 10). Creatinine clearance is estimated by the Cockcroft-Gault equation. The STEMI regimen continues for the hospital stay, up to 8 days, or until revascularization. SC, subcutaneous; IV, intravenous.

Kidney disease. Clearance of enoxaparin falls in proportion to creatinine clearance, and anti-Xa activity accumulates with repeated doses below 30 mL/min; the label therefore halves the dosing frequency. Patients on dialysis were excluded from the pivotal trials, and UFH is generally used in them. The 2018 American Society of Hematology (ASH) guideline on optimal anticoagulation suggests against adjusting LMWH doses by anti-Xa levels in patients with a creatinine clearance below 30 mL/min (conditional recommendation, very low certainty), preferring the label's renal dose reduction or a switch to UFH. The guideline cites poor assay standardization and a weak relation between anti-Xa levels and bleeding.

Obesity and low body weight. Enoxaparin is dosed by actual body weight. The same ASH guideline suggests dosing LMWH by actual body weight and not by a capped maximum dose, and against routine anti-Xa-guided adjustment in obesity, both as conditional recommendations based on very low certainty evidence. Data above about 150 kg are sparse; anti-Xa measurement after the third or fourth dose remains common practice at extremes of weight and in pregnancy despite these uncertainties. At the other extreme, women below 45 kg and men below 57 kg have higher exposure to fixed prophylactic doses and are observed closely for bleeding.

Older adults. Bleeding with enoxaparin rises with age, partly through unrecognized kidney impairment. In patients 75 years or older with STEMI treated by fibrinolysis, the omission of the intravenous bolus and the reduction to 0.75 mg/kg in ExTRACT-TIMI 25 were introduced after excess intracranial hemorrhage with full-dose regimens in older patients in earlier trials.

Switching between heparins. In SYNERGY, patients who were switched between enoxaparin and UFH around PCI had more bleeding than those who stayed on one agent. The agent started at admission is continued through the procedure, with intravenous enoxaparin supplementation timed to the last subcutaneous dose.

Trials of low-molecular-weight heparin against unfractionated heparin

The evidence for enoxaparin in cardiovascular disease comes from large randomized comparisons with UFH. In conservatively managed unstable angina, enoxaparin reduced recurrent ischemia. With an early invasive strategy, enoxaparin and UFH were equivalent for ischemic outcomes but enoxaparin caused more bleeding. With fibrinolysis for STEMI, enoxaparin given for the hospital stay was superior to 48 hours of UFH. Fondaparinux trials, which used enoxaparin or UFH as comparators, are described with fondaparinux below.

Randomized trials comparing low-molecular-weight heparin and unfractionated heparin (UFH). Rates are LMWH (or the first-named regimen) versus comparator. NSTE-ACS, non-ST-elevation acute coronary syndrome; TIMI, Thrombolysis in Myocardial Infarction.

The guideline consequences in ACS are set out in the 2025 ACC/AHA guideline. In NSTE-ACS, intravenous UFH is recommended (Class 1), and enoxaparin or fondaparinux is a recommended alternative when an early invasive approach is not anticipated (Class 1). Intravenous enoxaparin may be considered as an alternative to UFH at the time of PCI (Class 2b). In STEMI treated with fibrinolysis, parenteral anticoagulation continues for the hospital stay, up to 8 days, or until revascularization (Class 1), and enoxaparin is the recommended agent when no invasive approach is intended (Class 1). Management of each syndrome is described in the NSTE-ACS and STEMI hubs. For venous thromboembolism, LMWH and fondaparinux are preferred over intravenous UFH for initial parenteral therapy in most patients, and the choice of oral agent and duration are covered in the venous thromboembolism treatment hub and the pulmonary embolism hub.

Fondaparinux: pharmacology, dosing, and catheter thrombosis during PCI

Fondaparinux is a synthetic copy of the antithrombin-binding pentasaccharide, with a molecular weight of 1,728 daltons. It inhibits factor Xa through antithrombin and has no activity against thrombin. Subcutaneous bioavailability is complete, it binds no plasma protein other than antithrombin, its half-life is 17 to 21 hours, and it is excreted unchanged by the kidneys. These properties allow a single daily dose without monitoring and make it contraindicated with a creatinine clearance below 30 mL/min. No specific reversal agent exists, and protamine has no effect.

Prophylactic dosing is 2.5 mg once daily; prophylaxis is contraindicated in patients weighing less than 50 kg who are undergoing surgery. Treatment of venous thromboembolism is weight based: 5 mg daily below 50 kg, 7.5 mg from 50 to 100 kg, and 10 mg above 100 kg. In ACS, the dose is 2.5 mg once daily; with fibrinolysis for STEMI, the first dose is given intravenously.

OASIS-5. In 20,078 patients with NSTE-ACS, fondaparinux 2.5 mg daily was noninferior to enoxaparin 1 mg/kg twice daily for death, MI, or refractory ischemia at 9 days (5.8% versus 5.7%) and nearly halved major bleeding (2.2% versus 4.1%). Mortality at 30 and 180 days was lower with fondaparinux, a difference attributed to less bleeding. Among patients who underwent PCI, guiding-catheter thrombus was more frequent with fondaparinux (0.9% versus 0.4%).

OASIS-6. In 12,092 patients with STEMI, fondaparinux reduced death or reinfarction at 30 days from 11.2% to 9.7% compared with usual care. The benefit was confined to patients treated with fibrinolysis or no reperfusion, and there was no benefit in primary PCI, where catheter thrombosis again occurred more often.

FUTURA/OASIS-8. Because fondaparinux does not inhibit thrombin, it does not prevent contact activation of coagulation on catheters and guidewires. In 2,026 patients with NSTE-ACS treated with fondaparinux and undergoing PCI, low-dose UFH (50 U/kg) did not reduce the bleeding composite compared with standard-dose UFH (85 U/kg, or 60 U/kg with a glycoprotein IIb/IIIa inhibitor): 4.7% versus 5.8%. The secondary composite of bleeding and ischemic events was higher with the low dose (5.8% versus 3.9%; P=0.05). Standard-dose UFH is therefore given at PCI to patients receiving fondaparinux. The 2025 ACC/AHA ACS guideline states that fondaparinux should not be used to support PCI because of the risk of catheter thrombosis (Class 3: Harm).

Fondaparinux and HIT. Fondaparinux is too short to form the PF4 complexes that HIT antibodies recognize in most patients. Case reports of fondaparinux-associated HIT exist, but they are rare enough that the 2018 ASH guideline places patients receiving fondaparinux in the low-risk group, without platelet count monitoring. It is one of the suggested agents for acute HIT and is widely used for stable patients with normal kidney function.

Direct thrombin inhibitors: bivalirudin and argatroban

Bivalirudin. Bivalirudin is a 20-amino-acid synthetic analog of hirudin. It binds both the active site and exosite 1 of thrombin, but thrombin slowly cleaves it, so its inhibition is reversible. It inhibits free and fibrin-bound thrombin, does not bind plasma proteins or PF4, and has a linear dose-response. Its half-life is about 25 minutes with normal kidney function, rising to about 57 minutes in severe kidney impairment and about 3.5 hours on dialysis, because it undergoes glomerular filtration. It is labeled as the anticoagulant for patients undergoing PCI, including patients with HIT or HIT with thrombosis. Its use for treatment of acute HIT outside the catheterization laboratory is off label but supported by the 2018 ASH guideline. Its short half-life and independence from the liver make it the usual choice for HIT in patients with hepatic dysfunction, for HIT before urgent surgery or procedures, and for cardiopulmonary bypass in acute HIT.

Argatroban. Argatroban is a small synthetic molecule derived from L-arginine that binds reversibly to the active site of thrombin. It is metabolized in the liver, mainly by hydroxylation and aromatization, and excreted in bile; the US label states that metabolism by cytochrome P450 3A4/5 is not an important pathway in vivo. Its half-life is 39 to 51 minutes and is not changed by kidney disease, so no renal dose adjustment is needed. Hepatic impairment reduces clearance about fourfold. Argatroban is labeled for prophylaxis and treatment of thrombosis in HIT and for anticoagulation during PCI in patients with or at risk of HIT. In a prospective study with historical controls, argatroban reduced the composite of death, amputation, or new thrombosis in isolated HIT from 38.8% to 25.6%, with a nonsignificant reduction in HIT with thrombosis (56.5% to 43.8%), without more bleeding. Argatroban prolongs the prothrombin time and international normalized ratio (INR) more than other direct thrombin inhibitors at therapeutic concentrations, which complicates the transition to warfarin.

Dosing of the parenteral direct thrombin inhibitors, from the US prescribing information, the 2025 ACC/AHA ACS guideline, and published HIT protocols. Argatroban is supplied at 1 mg/mL in the standard premixed bag.

Other direct thrombin inhibitors have left the US market. Lepirudin, a recombinant hirudin that was approved for HIT, was discontinued in 2012, and desirudin is no longer marketed. Danaparoid, a mixture of heparan, dermatan, and chondroitin sulfates with predominant anti-Xa activity, is used for HIT in Europe, Canada, and elsewhere but is not available in the United States.

Procedural anticoagulation during PCI

Wire and catheter manipulation, balloon injury, and stent deployment expose tissue factor and foreign surfaces to blood, so every PCI is performed under systemic anticoagulation. The agents used are UFH, bivalirudin, and, less often in the United States, intravenous enoxaparin. Fondaparinux is never the sole procedural anticoagulant. Device and technique choices, including access and closure, are covered in the PCI technique hub.

Unfractionated heparin. UFH is the standard procedural anticoagulant. The 2021 ACC/AHA/SCAI revascularization guideline and the 2025 ACC/AHA ACS guideline both state that intravenous UFH is useful to reduce ischemic events in patients undergoing PCI (Class 1). Dosing is guided by the ACT. Without a glycoprotein IIb/IIIa inhibitor, the bolus is 70 to 100 U/kg, to an ACT of 250 to 300 seconds with the HemoTec device or 300 to 350 seconds with the Hemochron device. With a glycoprotein IIb/IIIa inhibitor, the bolus is 50 to 70 U/kg, to an ACT of 200 to 250 seconds. The 2025 ACS guideline uses an ACT target of 250 to 300 seconds, with additional UFH as needed in patients who already received an anticoagulant. Routine full-dose anticoagulation after uncomplicated PCI is not indicated.

Bivalirudin. Bivalirudin causes less bleeding than heparin combined with a glycoprotein IIb/IIIa inhibitor. Its advantage over heparin alone, in the era of radial access and selective glycoprotein IIb/IIIa inhibitor use, has varied with the comparator and with whether the infusion continues after PCI. Early trials found an excess of acute stent thrombosis with bivalirudin, attributed to its short half-life after the infusion stopped; BRIGHT-4 addressed this with a full-dose infusion for 2 to 4 hours after the procedure. The 2025 ACC/AHA ACS guideline states that bivalirudin is useful as an alternative to UFH in STEMI to reduce mortality and bleeding (Class 1), with the full-dose post-procedure infusion, and that it may be reasonable as an alternative to UFH in NSTE-ACS to reduce bleeding (Class 2b). The 2023 European Society of Cardiology ACS guideline gives bivalirudin with a full-dose post-PCI infusion a weaker Class 2a recommendation as an alternative to UFH in primary PCI.

Randomized trials of bivalirudin during PCI. Rates are bivalirudin versus heparin-based comparator. BARC, Bleeding Academic Research Consortium.

Enoxaparin. Intravenous enoxaparin at 0.5 mg/kg reduced bleeding compared with UFH in elective PCI (STEEPLE) and produced a nonsignificant reduction in the composite end point in primary PCI (ATOLL). Patients who arrive on subcutaneous enoxaparin receive no additional anticoagulant if the last dose was within 8 hours and 0.3 mg/kg intravenously if it was 8 to 12 hours earlier. Intravenous enoxaparin at PCI is a Class 2b alternative to UFH in the 2025 ACS guideline.

Glycoprotein IIb/IIIa inhibitors with heparin. When a glycoprotein IIb/IIIa inhibitor is used, the heparin dose and ACT target are lowered to limit bleeding. The 2025 ACC/AHA ACS guideline finds adjunctive intravenous or intracoronary glycoprotein IIb/IIIa inhibitors reasonable in PCI with a large thrombus burden, no-reflow, or slow flow (Class 2a), and recommends against routine use because of increased bleeding without ischemic benefit (Class 3: Harm).

Radial access and radial artery occlusion. Systemic heparin prevents radial artery occlusion after transradial catheterization. In SPIRIT OF ARTEMIS, in which 1,836 of 3,102 randomized patients underwent transradial diagnostic angiography without PCI, a heparin dose of 100 U/kg reduced early radial artery occlusion compared with 50 U/kg (3.0% versus 8.1%) without excess bleeding. The number needed to treat to prevent one occlusion was about 20. Patent hemostasis with the shortest effective compression complements anticoagulation. Weight-based doses of 50 to 100 U/kg are used in diagnostic radial procedures.

HIT and PCI. In patients with HIT undergoing PCI, the 2021 revascularization guideline states that bivalirudin or argatroban should be used in place of UFH to avoid thrombotic complications (Class 1). The 2018 ASH guideline suggests bivalirudin for PCI in acute or subacute HIT, with argatroban as a substitute when bivalirudin is unavailable. It also suggests bivalirudin in preference to UFH even in remote HIT, with heparin acceptable only when no suitable alternative exists and limited to the procedure.

Neuraxial anesthesia and other procedures during heparin therapy

Spinal or epidural hematoma after neuraxial anesthesia or lumbar puncture is rare but can cause permanent paralysis, and the enoxaparin and fondaparinux labels carry a boxed warning. Under the enoxaparin label, placement or removal of a neuraxial catheter is delayed at least 12 hours after prophylactic doses and at least 24 hours after treatment doses, with longer intervals in kidney impairment. The 2018 American Society of Regional Anesthesia and Pain Medicine (ASRA) guidelines advise removing an indwelling neuraxial catheter 4 to 6 hours after the last dose of intravenous heparin, after the coagulation status is assessed, and restarting heparin no sooner than 1 hour after removal. Neuraxial techniques are generally avoided during therapeutic fondaparinux and direct thrombin inhibitor therapy because of their long or unmonitorable effect. The timing of anticoagulant interruption before surgery and cardiac procedures is covered in the periprocedural hub.

Heparin and LMWH in pregnancy

Neither UFH nor LMWH crosses the placenta, and neither is secreted in clinically important amounts in breast milk, so both are safe for the fetus and during lactation. LMWH is preferred for most indications in pregnancy because of its predictable dosing, lower risk of HIT, and less bone loss. The plasma volume, glomerular filtration, and volume of distribution increase during pregnancy, so LMWH doses based on early-pregnancy weight can become subtherapeutic, and peak anti-Xa monitoring is used for high-risk indications. LMWH is often changed to UFH near term, or held for a planned delivery, to allow neuraxial anesthesia within the intervals given for neuraxial procedures. Fondaparinux is reserved for women with HIT or severe heparin allergy, and argatroban and bivalirudin have limited data in pregnancy. Women with mechanical heart valves need the highest anticoagulant intensity in pregnancy; dose-adjusted LMWH with peak and trough anti-Xa targets, UFH, and warfarin are compared in the prosthetic heart valve hub.

Heparin for cardiopulmonary bypass and extracorporeal circuits

Cardiopulmonary bypass exposes blood to a large synthetic surface and requires the most intense anticoagulation in clinical practice. UFH is used for nearly every case because it can be given in very large doses, monitored at the bedside by the ACT, and fully reversed with protamine at the end of bypass. A bolus of 300 to 400 U/kg is given before cannulation, heparin is added to the circuit prime, and further doses maintain an ACT, most commonly 480 seconds or more; some centers titrate to a measured heparin concentration. Heparin resistance on bypass, usually from antithrombin deficiency after preoperative heparin, is treated with additional heparin and antithrombin concentrate. After protamine, a delayed return of heparin effect (heparin rebound) can cause bleeding and is treated with small additional protamine doses.

Extracorporeal membrane oxygenation (ECMO) and temporary ventricular assist devices also run on continuous UFH, at much lower intensity, titrated to anti-Xa activity, aPTT, or ACT by device and center protocol. Thrombocytopenia during mechanical support is common and seldom caused by HIT, but the diagnosis is difficult because consumption and hemolysis lower the platelet count. Device-specific anticoagulation and its complications are covered in the temporary mechanical circulatory support hub and the venoarterial ECMO hub.

Pathophysiology of heparin-induced thrombocytopenia

HIT is an adverse drug reaction mediated by immunoglobulin G (IgG) antibodies that recognize complexes of PF4 and heparin. PF4 is a positively charged chemokine stored in platelet alpha granules and released on activation. When it binds a long, negatively charged heparin chain at the right stoichiometric ratio, PF4 tetramers cluster into large multimolecular complexes and expose new epitopes (neoepitopes). In susceptible patients these complexes provoke an antibody response that resembles a secondary immune response: IgG antibodies appear within about 5 to 10 days of a first exposure, without a preceding IgM phase.

The pathogenic antibodies bind PF4-heparin complexes on the platelet surface, and their Fc portions then cross-link the platelet FcγRIIa receptors. Platelets are activated, release more PF4, and shed procoagulant microparticles. The antibodies also bind to PF4 attached to endothelial glycosaminoglycans and activate monocytes, which express tissue factor, and neutrophils, which undergo NETosis. The net result is massive thrombin generation. Thrombocytopenia reflects consumption of activated platelets and their clearance by the reticuloendothelial system; bleeding is unusual despite counts that may fall to 20 to 50 × 10⁹/L, and the dominant clinical risk is thrombosis.

Only a minority of patients who form anti-PF4/heparin antibodies develop HIT. Antibodies of the IgM and IgA classes, and IgG antibodies that do not activate platelets, are detected by immunoassays but cause no disease. After cardiac surgery, PF4/heparin antibodies are detectable in a large proportion of patients without HIT. In a prospective study of 744 patients, the frequency of heparin-dependent IgG antibodies by antigen assay was 50% after cardiac surgery with UFH, compared with 7.5% after orthopedic surgery with LMWH. Among patients who formed antibodies, however, orthopedic patients were much more likely to develop HIT. This dissociation between seroconversion and disease is the basis of the diagnostic algorithm: a positive immunoassay alone does not establish HIT.

LMWH forms smaller complexes with PF4 and is less immunogenic, which is why HIT is about 10 times less frequent with LMWH than with UFH. Once antibodies have formed, however, they cross-react with PF4-LMWH complexes in most patients, so LMWH is contraindicated in acute HIT.

Incidence and risk factors for HIT

The frequency of HIT ranges from less than 0.1% to about 7% of hospitalized patients exposed to heparin. The 2018 ASH guideline identifies three determinants: the type of heparin (UFH about 10 times the risk of LMWH, fondaparinux negligible), the patient population (surgical and major trauma patients about three times the risk of medical patients), and the duration of exposure (2% to 3% of patients receiving UFH for 6 or more days). A meta-analysis of thromboprophylaxis trials, mostly after orthopedic surgery, found HIT in 2.6% of patients receiving UFH and 0.2% of those receiving LMWH. In a randomized trial of UFH and LMWH prophylaxis after hip surgery, HIT occurred in 2.7% of patients given UFH and none given LMWH, and 8 of the 9 patients with HIT had thrombosis. Women are affected more often than men. HIT is uncommon in children, in pregnancy, and after very brief exposure such as catheter flushes, but it can occur.

Risk of HIT by population and the platelet count monitoring suggested by the 2018 ASH guideline (conditional recommendations). Patients after cardiac surgery with cardiopulmonary bypass are in the high-risk group; the risk of HIT is 1% to 3% when UFH is continued beyond the first postoperative week.

Clinical presentation, complications, and outcomes of HIT

Typical-onset HIT. The platelet count begins to fall 5 to 10 days after the start of heparin (the first day of heparin is day 0). The fall is usually more than 50% of the postheparin peak, the nadir is usually moderate, and a nadir below 20 × 10⁹/L is unusual and suggests another cause or severe autoimmune HIT. In patients whose count was high after surgery, HIT can occur while the absolute count is still normal; the relevant reference is the highest count after the start of heparin, not the preoperative value. In a series of 243 patients with serologically confirmed HIT, 70% had a typical onset 4 or more days after the start of heparin, and prior heparin exposure did not change that timing.

Rapid-onset HIT. The remaining 30% of patients in that series developed thrombocytopenia within hours (median 10.5 hours) of starting heparin. All had received heparin within the previous 100 days, and the reaction reflects circulating antibodies from the recent exposure, not an anamnestic response. A rapid fall in the platelet count after starting heparin is attributed to HIT only when heparin was given within the previous 100 days, most strongly within the previous 30 days.

Delayed-onset HIT. Thrombocytopenia and thrombosis can begin or worsen days after heparin has been stopped, typically within 3 weeks, in patients with strongly platelet-activating antibodies. Delayed-onset HIT commonly presents as a readmission with thrombosis after a recent hospital stay, often after cardiac surgery or orthopedic surgery. Giving heparin for the new thrombosis worsens the course; the platelet count falls further and the thrombosis progresses.

Thrombosis. Thrombosis complicates one-third to one-half of HIT episodes. In a 14-year series of 127 patients, venous thrombosis was four times as common as arterial thrombosis, pulmonary embolism occurred in 25%, and about half the patients were recognized only after a thrombotic event. Among the 62 patients first recognized with isolated thrombocytopenia, the 30-day risk of thrombosis was 52.8%, whether heparin was stopped or warfarin was substituted. Stopping heparin alone is therefore inadequate treatment. Venous events include deep vein thrombosis (often bilateral, often in the arm with a central catheter), pulmonary embolism, cerebral venous sinus thrombosis, and adrenal vein thrombosis with hemorrhagic adrenal infarction and adrenal insufficiency. Arterial events include limb artery occlusion by platelet-rich "white clot," stroke, myocardial infarction, mesenteric and renal artery thrombosis, and thrombosis of grafts, valves, and devices.

Skin and systemic manifestations. Painful erythematous plaques or skin necrosis at subcutaneous heparin injection sites occur in a minority, often without thrombocytopenia, and count as a manifestation of HIT. An acute systemic reaction within 30 minutes of an intravenous heparin bolus (fever, chills, hypertension, tachycardia, dyspnea, chest pain, and transient global amnesia or cardiorespiratory arrest) reflects sudden platelet activation by circulating antibodies and is accompanied by an abrupt fall in the platelet count.

Disseminated intravascular coagulation and venous limb gangrene. Severe HIT can produce overt disseminated intravascular coagulation with low fibrinogen and a prolonged INR. Venous limb gangrene, acral necrosis distal to a deep vein thrombosis despite palpable pulses, occurs when warfarin is started during acute HIT: the rapid fall in protein C, combined with continuing thrombin generation, causes microvascular thrombosis. In the series that defined the syndrome, it was associated with a supratherapeutic INR, a marker of severe depletion of factor VII in parallel with protein C.

Outcomes. The 2018 ASH guideline cites amputation in about 1% to 3% and death in about 5% to 10% of patients, rates that fall with prompt recognition and treatment.

Differential diagnosis of thrombocytopenia in hospitalized patients receiving heparin

Thrombocytopenia is common in hospitalized patients, and most thrombocytopenia during heparin therapy is not HIT. Overdiagnosis exposes patients to costly anticoagulants with higher bleeding risk and to permanent mislabeling of a heparin allergy, while underdiagnosis exposes them to thrombosis. The timing of the fall, its magnitude, the nadir, and the presence of a competing explanation are weighed together.

Differential diagnosis of thrombocytopenia during heparin therapy. HIT is also distinguished from vaccine-induced immune thrombotic thrombocytopenia and spontaneous HIT, which occur without heparin exposure.

The 4Ts score for heparin-induced thrombocytopenia

The 4Ts score estimates the pretest probability of HIT from four clinical features, each scored 0, 1, or 2: the magnitude of thrombocytopenia, the timing of the fall, thrombosis or other sequelae, and the presence of other causes. A total of 0 to 3 indicates low probability, 4 to 5 intermediate, and 6 to 8 high. The 2018 ASH guideline recommends the 4Ts score over gestalt for estimating pretest probability (strong recommendation).

In a meta-analysis of 13 studies with 3,068 patients, 56% of patients with suspected HIT had a low-probability score, and the negative predictive value of a low score was 0.998. The positive predictive value was 0.14 for an intermediate score and 0.64 for a high score. A low score therefore excludes HIT reliably regardless of who scores it, whereas intermediate and high scores identify patients who need laboratory testing. Interobserver agreement is imperfect, mainly for the "other causes" item, and the score performs less well after cardiac surgery and in the intensive care unit, where thrombocytopenia from other causes is almost universal.

The 4Ts score. Day 0 is the first day of the most recent heparin exposure. The fall is calculated from the highest platelet count after the start of heparin.

Scoring from the original 4Ts description; predictive values from the 2012 meta-analysis; testing and management sequence from the 2018 ASH guideline. NPV, negative predictive value; PPV, positive predictive value; SRA, serotonin-release assay; HIPA, heparin-induced platelet activation assay.

The HIT Expert Probability (HEP) score is an alternative that includes more items, weighted by expert opinion. Neither score replaces laboratory testing when the probability is intermediate or high.

Laboratory testing for HIT: immunoassays and functional assays

Immunoassays. Immunoassays detect antibodies to PF4-polyanion complexes. Enzyme-linked immunosorbent assays (ELISA) report an optical density (OD), and rapid automated assays (chemiluminescent, latex agglutination, particle gel, and lateral flow) give results within an hour. IgG-specific assays are more specific than polyspecific assays that also detect IgM and IgA. Sensitivity is high, above 95% for most assays, so a negative result excludes HIT in most patients. Specificity is limited because nonpathogenic antibodies are common, particularly after cardiac surgery. The magnitude of the reaction carries information: the higher the OD, the higher the probability that the antibodies activate platelets, and a weakly positive ELISA (OD between about 0.4 and 1.0) in a patient with an intermediate score is more often a false positive than a true one.

Functional assays. Functional assays measure the ability of patient serum to activate donor platelets in the presence of heparin. The serotonin-release assay (SRA) measures release of radiolabeled serotonin from washed platelets at a therapeutic heparin concentration and its suppression at a high concentration; the heparin-induced platelet activation (HIPA) assay measures platelet aggregation in microtiter wells. Both have high sensitivity and specificity in experienced laboratories and are the reference standard, but they are performed in few laboratories, take days to return, and cannot guide the initial decision. A negative functional assay with a positive immunoassay usually indicates nonpathogenic antibodies; false-negative functional assays occur when antibody levels are low or when samples are handled poorly.

Other tests. Platelet aggregation studies with heparin have low sensitivity and are no longer recommended. Flow cytometric assays of platelet activation and assays that add PF4 without heparin (used for autoimmune HIT and vaccine-induced thrombotic thrombocytopenia) are available in reference laboratories.

The 2018 ASH guideline combines the 4Ts score and the test results into a single sequence. With a low score, no testing and no empiric treatment are recommended (strong recommendations), because a positive immunoassay in that setting is far more likely to be false than true. With an intermediate or high score, an immunoassay is recommended (strong), followed by a functional assay when the immunoassay is positive and a functional assay is available (conditional).

Initial management of suspected HIT by 4Ts score and immunoassay result, from the 2018 ASH guideline (recommendations 2.2 to 2.10). With a high score and a negative immunoassay, false-negative results are rare, and repeat or alternative testing is reasonable when suspicion remains strong.

Treatment of acute HIT

Withdrawal of all heparin. Every source of heparin is stopped when HIT is suspected at an intermediate or high score: therapeutic and prophylactic UFH and LMWH, heparin flushes of arterial and venous catheters, heparin locks, heparin-coated catheters (including some pulmonary artery catheters), heparin in dialysis circuits, and heparin-containing prothrombin complex concentrates. The patient's record is flagged as HIT, with the date of the episode, and not as a heparin allergy.

A therapeutic non-heparin anticoagulant. Stopping heparin alone leaves a thrombotic risk of about 50% in the following month. The 2018 ASH guideline recommends a non-heparin anticoagulant (strong) and suggests argatroban, bivalirudin, danaparoid, fondaparinux, or a direct oral anticoagulant (conditional), at therapeutic and not prophylactic intensity (strong). The choice among them depends on clinical stability, bleeding risk, the likelihood of an urgent procedure, kidney and liver function, cost, and local experience. No randomized trial compares the agents.

Non-heparin anticoagulants for acute HIT. DIC, disseminated intravascular coagulation.

Direct oral anticoagulants. A prospective study and later reviews of rivaroxaban, apixaban, dabigatran, and edoxaban in serologically confirmed HIT found low rates of new thrombosis and major bleeding when a DOAC was used as primary therapy or after initial parenteral treatment. In the Hamilton series, rivaroxaban was given in the regimen used for venous thromboembolism (15 mg twice daily, followed by 20 mg once daily). DOACs avoid the INR problems of argatroban and warfarin and allow discharge without injections, which favors them for stable patients. Patients with life- or limb-threatening thrombosis, those likely to need an urgent procedure, and those with severe kidney impairment are usually treated with a parenteral direct thrombin inhibitor first.

Monitoring direct thrombin inhibitors in HIT. HIT with disseminated intravascular coagulation prolongs the baseline aPTT, so an aPTT in the target range may reflect the coagulopathy and not the drug, and the patient may be underanticoagulated (aPTT confounding). Thrombosis that progresses despite a therapeutic aPTT suggests this mechanism. A drug-calibrated dilute thrombin time or ecarin-based assay, where available, measures drug concentration; a switch to an agent that is not titrated by aPTT is the alternative.

Platelet transfusion. Platelet transfusion was long avoided because of a theoretical risk of fueling thrombosis. The 2018 ASH guideline suggests against routine platelet transfusion in patients at average bleeding risk (conditional) and notes that it may be an option for patients with active bleeding or at high risk of bleeding. Bleeding is uncommon in HIT despite low counts.

Antiplatelet agents, filters, and screening ultrasound. The ASH guideline suggests using the non-heparin anticoagulant alone, without an added antiplatelet agent unless another indication exists (conditional), and recommends against routine inferior vena cava filters (strong). Because many patients with apparently isolated HIT already have asymptomatic deep vein thrombosis, the guideline suggests bilateral lower-extremity compression ultrasonography in isolated HIT, and ultrasound of an upper limb with a central venous catheter (conditional). A thrombus found this way converts the diagnosis to HIT with thrombosis and lengthens treatment.

Refractory HIT: intravenous immunoglobulin and plasma exchange. In severe or persisting HIT, including autoimmune forms with platelet counts that do not recover, high-dose intravenous immunoglobulin (1 g/kg daily for 2 days) inhibits FcγRIIa-mediated platelet activation and raises the platelet count within days. Therapeutic plasma exchange removes the antibodies and is used mainly before cardiac surgery in acute or subacute HIT. Both uses are off label and rest on case series.

Transition to oral anticoagulation and duration of treatment in HIT

Warfarin. Warfarin is not started during acute HIT. The 2018 ASH guideline recommends against starting a vitamin K antagonist until the platelet count has recovered, usually to at least 150 × 10⁹/L (strong). When HIT is diagnosed in a patient already taking warfarin, warfarin is stopped and intravenous vitamin K is given, while the non-heparin anticoagulant is started. Once the count has recovered, warfarin is started at an expected maintenance dose without a loading dose and overlapped with the parenteral agent for at least 5 days and until the INR is therapeutic.

The argatroban-warfarin overlap. Argatroban raises the INR, so a combined INR overstates the warfarin effect. Under the US label, when argatroban is given at up to 2 µg/kg/min, it can be stopped when the INR on combined therapy exceeds 4; the INR is repeated 4 to 6 hours later, and argatroban is resumed if the INR has fallen below the therapeutic range, with the process repeated daily. At argatroban doses above 2 µg/kg/min, the dose is temporarily reduced to 2 µg/kg/min and the INR measured 4 to 6 hours later. A chromogenic factor X activity, which is not affected by argatroban, measures the warfarin effect without interference and is used where available, with laboratory-specific thresholds. Bivalirudin raises the INR less, and fondaparinux and DOACs do not interfere.

DOACs after HIT. A DOAC can follow any parenteral agent without overlap: it is started when the infusion is stopped (argatroban or bivalirudin) or when the next fondaparinux dose would be due. For patients with subacute HIT A (platelets recovered, functional assay still positive) who need continued anticoagulation, the ASH guideline suggests a DOAC over a vitamin K antagonist (conditional).

Duration. In isolated HIT without thrombosis, the 2018 ASH guideline suggests anticoagulation at least until platelet recovery and not for 3 months or longer unless the platelet count fails to recover (conditional). Many centers treat isolated HIT for up to 4 weeks, a duration noted in the 2012 CHEST guideline, because the excess thrombotic risk persists for weeks after heparin is stopped. HIT with thrombosis is treated as provoked thrombosis, for 3 months in the 2012 CHEST guideline and typically 3 to 6 months in the 2018 ASH guideline. The duration of anticoagulation for other venous thromboembolism is covered in the venous thromboembolism treatment hub.

Phases of HIT and re-exposure to heparin

HIT antibodies are transient. In the serial study of 243 patients, heparin-dependent antibodies became undetectable at a median of 50 days by functional assay and 85 days by immunoassay, and 7 patients given heparin again after their antibodies had disappeared did not develop recurrent HIT. Heparin does not reliably restimulate antibody formation within the first days of re-exposure. This natural history is the basis of the phases used by the 2018 ASH guideline to decide whether heparin can be given for a procedure.

Phases of HIT defined in the 2018 ASH guideline. A suspected case confirmed by the 4Ts score and an immunoassay alone (without a functional assay) is classified by the immunoassay and the platelet count.

For treatment or prophylaxis of venous thromboembolism in a patient with remote HIT, the ASH guideline recommends a non-heparin anticoagulant (apixaban, dabigatran, danaparoid, edoxaban, fondaparinux, rivaroxaban, or a vitamin K antagonist) over UFH or LMWH (strong). It suggests an emergency identifier for patients whose HIT was within the previous 3 months, and not for older episodes. Brief heparin exposure is confined to procedures (cardiac surgery and, when no alternative exists, catheterization), with platelet counts monitored afterward because delayed-onset HIT can appear 5 to 10 days later.

Cardiac surgery, catheterization, and dialysis in patients with HIT

Cardiac surgery is the setting in which HIT is most often suspected and in which the choice of anticoagulant has the largest consequences, because no alternative matches heparin with protamine for cardiopulmonary bypass. The 2018 ASH guideline separates patients by phase.

Acute or subacute HIT A. When surgery can wait, it is delayed until the patient reaches subacute HIT B or remote HIT (good practice statement), which usually takes weeks. When it cannot wait, the guideline suggests one of three approaches (conditional): intraoperative bivalirudin; preoperative or intraoperative plasma exchange followed by intraoperative heparin; or intraoperative heparin combined with a potent antiplatelet agent (prostacyclin analogs or tirofiban). Bivalirudin for bypass is given as a weight-based bolus, a dose added to the pump prime, and a continuous infusion, monitored by ACT. These regimens come from small open-label studies and HIT cohorts; one of the studies, EVOLUTION-ON, compared bivalirudin with heparin in patients without HIT. Bivalirudin requires modified perfusion technique, because thrombin cleaves the drug and blood clots in stagnant areas of the circuit, in cardiotomy suction, and in the operative field. No reversal agent exists at the end of bypass; the short half-life and elimination by hemofiltration bring the effect to an end.

Subacute HIT B and remote HIT. Because platelet-activating antibodies are gone, the ASH guideline suggests intraoperative heparin, limited to the operation, over non-heparin anticoagulants (conditional). All postoperative heparin, including flushes, is avoided, and a non-heparin agent is used for postoperative prophylaxis or treatment. Platelet counts are followed because delayed-onset HIT can follow the re-exposure. After cardiac surgery with heparin, 25% to 50% of patients form anti-PF4/heparin antibodies within 5 to 10 days without disease, so a positive immunoassay in that window is interpreted alongside the timing of the platelet fall and a functional assay.

Cardiac catheterization and PCI. In acute or subacute HIT A, the ASH guideline suggests bivalirudin for PCI, with argatroban when bivalirudin is unavailable (conditional). In subacute HIT B and remote HIT, it suggests bivalirudin over UFH (conditional), and UFH is acceptable only when no alternative is available and only during the procedure. All heparin flushes and heparin-coated devices are avoided. Argatroban is labeled for PCI in HIT at a 350 µg/kg bolus and 25 µg/kg/min, to an ACT of 300 to 450 seconds, and bivalirudin at its standard PCI dose.

Renal replacement therapy. In acute HIT, the ASH guideline suggests argatroban, danaparoid, or bivalirudin for anticoagulation of the dialysis circuit (conditional). In subacute or remote HIT in patients not otherwise anticoagulated, it suggests regional citrate over heparin or other agents (conditional); citrate is not appropriate in acute HIT, which requires systemic anticoagulation.

ECMO and mechanical circulatory support. In patients with HIT on ECMO or a temporary ventricular assist device, bivalirudin and argatroban replace heparin in the circuit and the patient, with ACT, aPTT, or drug-level monitoring by center protocol. Stagnant zones in the circuit clot more readily with direct thrombin inhibitors, and the circuit is inspected and replaced as needed. The diagnosis itself is frequently uncertain on mechanical support, and a functional assay is particularly important before a patient is labeled.

Autoimmune HIT, spontaneous HIT, and vaccine-induced immune thrombotic thrombocytopenia

Some anti-PF4 antibodies activate platelets even without heparin. Autoimmune HIT describes the resulting syndromes: delayed-onset HIT; persisting (refractory) HIT, in which thrombocytopenia lasts a week or more after heparin is stopped; HIT induced by a heparin flush alone; fondaparinux-associated HIT; and severe HIT with platelet counts below 20 × 10⁹/L and disseminated intravascular coagulation. Serum from these patients activates platelets in a functional assay without added heparin. Treatment uses agents that are not titrated by aPTT (fondaparinux, a DOAC, or danaparoid), because aPTT-adjusted direct thrombin inhibitors are frequently underdosed in coagulopathy. High-dose intravenous immunoglobulin is used to interrupt platelet activation.

Spontaneous HIT syndrome is a HIT-like disorder that occurs without any heparin exposure, most often after total knee replacement or infection, and presents with thrombocytopenia and thrombosis. The PF4/heparin immunoassay and functional assays are strongly positive. It is treated as autoimmune HIT, with strict avoidance of heparin.

Vaccine-induced immune thrombotic thrombocytopenia (VITT) is a related disorder of platelet-activating anti-PF4 antibodies that followed adenoviral vector vaccines against COVID-19. In the first reported series, patients presented 5 to 16 days after vaccination with thrombocytopenia, thrombosis at unusual sites (cerebral venous sinus and splanchnic veins), and disseminated intravascular coagulation, without heparin exposure. VITT antibodies bind PF4 at a different site from HIT antibodies and are detected by ELISA but often missed by rapid immunoassays. Treatment is non-heparin anticoagulation with intravenous immunoglobulin. VITT-like disorders also occur rarely after adenovirus infection.

Errors in parenteral anticoagulation and HIT management

• A therapeutic aPTT range carried over from another hospital or another reagent: aPTT ranges are reagent specific and calibrated to an anti-Xa level of 0.3 to 0.7 IU/mL.
• Escalating heparin for a short aPTT in an acutely ill patient without measuring anti-Xa activity: high factor VIII and fibrinogen shorten the aPTT while the heparin level is therapeutic, and escalation increases bleeding.
• Heparin anti-Xa monitoring in a patient recently switched from apixaban or rivaroxaban: the residual DOAC raises the anti-Xa result and leads to underdosing of heparin.
• Full-dose enoxaparin every 12 hours with a creatinine clearance below 30 mL/min, or enoxaparin in a patient on dialysis: anti-Xa activity accumulates and bleeding increases.
• Switching between enoxaparin and UFH around PCI, or giving UFH for PCI within 12 hours of a therapeutic subcutaneous enoxaparin dose (Class 3: Harm in the 2021 ACC/AHA/SCAI revascularization guideline): bleeding increases.
• Fondaparinux alone during PCI: catheter thrombosis occurs, and standard-dose UFH is given at the procedure.
• Stopping a bivalirudin infusion immediately after primary PCI: the excess of acute stent thrombosis in early trials was addressed by a full-dose infusion for 2 to 4 hours afterward.
• Ordering a PF4/heparin immunoassay in a patient with a low 4Ts score: false-positive results lead to unnecessary non-heparin anticoagulation and a permanent heparin allergy label.
• Stopping heparin without starting a non-heparin anticoagulant: half of patients with isolated HIT develop thrombosis within 30 days.
• Switching from UFH to LMWH in suspected HIT: the antibodies cross-react with LMWH in most patients.
• Starting warfarin before platelet recovery, or continuing it when HIT is diagnosed: protein C depletion causes venous limb gangrene and skin necrosis.
• Stopping argatroban when the combined INR first reaches 2 to 3: the warfarin effect alone is still subtherapeutic.
• Overlooking heparin flushes, heparin-coated catheters, and heparin in dialysis circuits after the diagnosis: continued exposure sustains platelet activation.
• Calculating the platelet fall from the preoperative value after cardiac surgery: the 50% fall is measured from the postoperative peak.

References

1. Cuker A, Arepally GM, Chong BH, et al. American Society of Hematology 2018 guidelines for management of venous thromboembolism: heparin-induced thrombocytopenia. Blood Adv. 2018;2:3360–3392. doi:10.1182/bloodadvances.2018024489
2. Rao SV, O'Donoghue ML, Ruel M, et al. 2025 ACC/AHA/ACEP/NAEMSP/SCAI guideline for the management of patients with acute coronary syndromes: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2025;151:e771–e862 (corrections: 2025;151:e865, 2025;151:e1098, and 2025;152:e402). doi:10.1161/CIR.0000000000001309
3. Lawton JS, Tamis-Holland JE, Bangalore S, et al. 2021 ACC/AHA/SCAI guideline for coronary artery revascularization: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2022;145:e18–e114 (corrections: 2022;145:e772 and 2024;150:e105). doi:10.1161/CIR.0000000000001038
4. Levine GN, Bates ER, Blankenship JC, et al. 2011 ACCF/AHA/SCAI guideline for percutaneous coronary intervention. Circulation. 2011;124:e574–e651 (correction: 2012;125:e412). doi:10.1161/CIR.0b013e31823ba622
5. Stevens SM, Woller SC, Baumann Kreuziger L, et al. Antithrombotic therapy for VTE disease: second update of the CHEST guideline and expert panel report. Chest. 2021;160:e545–e608 (erratum: 2022;162:269). doi:10.1016/j.chest.2021.07.055
6. Witt DM, Nieuwlaat R, Clark NP, et al. American Society of Hematology 2018 guidelines for management of venous thromboembolism: optimal management of anticoagulation therapy. Blood Adv. 2018;2:3257–3291. doi:10.1182/bloodadvances.2018024893
7. Garcia DA, Baglin TP, Weitz JI, Samama MM. Parenteral anticoagulants: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2012;141(2 Suppl):e24S–e43S (errata: 2012;141:1369 and 2013;144:721). doi:10.1378/chest.11-2291
8. Linkins LA, Dans AL, Moores LK, et al. Treatment and prevention of heparin-induced thrombocytopenia: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2012;141(2 Suppl):e495S–e530S (erratum: 2015;148:1529). doi:10.1378/chest.11-2303
9. Horlocker TT, Vandermeuelen E, Kopp SL, et al. Regional anesthesia in the patient receiving antithrombotic or thrombolytic therapy: American Society of Regional Anesthesia and Pain Medicine evidence-based guidelines (fourth edition). Reg Anesth Pain Med. 2018;43:263–309 (erratum: 2018;43:566). doi:10.1097/AAP.0000000000000763
10. Raschke RA, Reilly BM, Guidry JR, et al. The weight-based heparin dosing nomogram compared with a “standard care” nomogram: a randomized controlled trial. Ann Intern Med. 1993;119:874–881. doi:10.7326/0003-4819-119-9-199311010-00002
11. Levine MN, Hirsh J, Gent M, et al. A randomized trial comparing activated thromboplastin time with heparin assay in patients with acute venous thromboembolism requiring large daily doses of heparin. Arch Intern Med. 1994;154:49–56.
12. Levy JH, Connors JM. Heparin resistance: clinical perspectives and management strategies. N Engl J Med. 2021;385:826–832. doi:10.1056/NEJMra2104091
13. Levy JH, Sniecinski RM, Maier CL, et al. Finding a common definition of heparin resistance in adult cardiac surgery: communication from the ISTH SSC subcommittee on perioperative and critical care thrombosis and hemostasis. J Thromb Haemost. 2024;22:1249–1257. doi:10.1016/j.jtha.2024.01.001
14. Kearon C, Ginsberg JS, Julian JA, et al. Comparison of fixed-dose weight-adjusted unfractionated heparin and low-molecular-weight heparin for acute treatment of venous thromboembolism. JAMA. 2006;296:935–942. doi:10.1001/jama.296.8.935
15. Cohen M, Demers C, Gurfinkel EP, et al. A comparison of low-molecular-weight heparin with unfractionated heparin for unstable coronary artery disease (ESSENCE). N Engl J Med. 1997;337:447–452. doi:10.1056/NEJM199708143370702
16. Ferguson JJ, Califf RM, Antman EM, et al. Enoxaparin vs unfractionated heparin in high-risk patients with non-ST-segment elevation acute coronary syndromes managed with an intended early invasive strategy: primary results of the SYNERGY randomized trial. JAMA. 2004;292:45–54. doi:10.1001/jama.292.1.45
17. Antman EM, Morrow DA, McCabe CH, et al. Enoxaparin versus unfractionated heparin with fibrinolysis for ST-elevation myocardial infarction (ExTRACT-TIMI 25). N Engl J Med. 2006;354:1477–1488. doi:10.1056/NEJMoa060898
18. Montalescot G, White HD, Gallo R, et al. Enoxaparin versus unfractionated heparin in elective percutaneous coronary intervention (STEEPLE). N Engl J Med. 2006;355:1006–1017. doi:10.1056/NEJMoa052711
19. Montalescot G, Zeymer U, Silvain J, et al. Intravenous enoxaparin or unfractionated heparin in primary percutaneous coronary intervention for ST-elevation myocardial infarction: the international randomised open-label ATOLL trial. Lancet. 2011;378:693–703. doi:10.1016/S0140-6736(11)60876-3
20. Yusuf S, Mehta SR, Chrolavicius S, et al. Comparison of fondaparinux and enoxaparin in acute coronary syndromes (OASIS-5). N Engl J Med. 2006;354:1464–1476. doi:10.1056/NEJMoa055443
21. Yusuf S, Mehta SR, Chrolavicius S, et al. Effects of fondaparinux on mortality and reinfarction in patients with acute ST-segment elevation myocardial infarction: the OASIS-6 randomized trial. JAMA. 2006;295:1519–1530. doi:10.1001/jama.295.13.joc60038
22. Steg PG, Jolly SS, Mehta SR, et al. Low-dose vs standard-dose unfractionated heparin for percutaneous coronary intervention in acute coronary syndromes treated with fondaparinux: the FUTURA/OASIS-8 randomized trial. JAMA. 2010;304:1339–1349. doi:10.1001/jama.2010.1320
23. Stone GW, McLaurin BT, Cox DA, et al. Bivalirudin for patients with acute coronary syndromes (ACUITY). N Engl J Med. 2006;355:2203–2216. doi:10.1056/NEJMoa062437
24. Stone GW, Witzenbichler B, Guagliumi G, et al. Bivalirudin during primary PCI in acute myocardial infarction (HORIZONS-AMI). N Engl J Med. 2008;358:2218–2230. doi:10.1056/NEJMoa0708191
25. Steg PG, van 't Hof A, Hamm CW, et al. Bivalirudin started during emergency transport for primary PCI (EUROMAX). N Engl J Med. 2013;369:2207–2217. doi:10.1056/NEJMoa1311096
26. Shahzad A, Kemp I, Mars C, et al. Unfractionated heparin versus bivalirudin in primary percutaneous coronary intervention (HEAT-PPCI): an open-label, single centre, randomised controlled trial. Lancet. 2014;384:1849–1858. doi:10.1016/S0140-6736(14)60924-7
27. Valgimigli M, Frigoli E, Leonardi S, et al. Bivalirudin or unfractionated heparin in acute coronary syndromes (MATRIX). N Engl J Med. 2015;373:997–1009. doi:10.1056/NEJMoa1507854
28. Erlinge D, Omerovic E, Fröbert O, et al. Bivalirudin versus heparin monotherapy in myocardial infarction (VALIDATE-SWEDEHEART). N Engl J Med. 2017;377:1132–1142. doi:10.1056/NEJMoa1706443
29. Li Y, Liang Z, Qin L, et al. Bivalirudin plus a high-dose infusion versus heparin monotherapy in patients with ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention: a randomised trial (BRIGHT-4). Lancet. 2022;400:1847–1857. doi:10.1016/S0140-6736(22)01999-7
30. Hahalis GN, Leopoulou M, Tsigkas G, et al. Multicenter randomized evaluation of high versus standard heparin dose on incident radial arterial occlusion after transradial coronary angiography: the SPIRIT OF ARTEMIS study. JACC Cardiovasc Interv. 2018;11:2241–2250. doi:10.1016/j.jcin.2018.08.009
31. Warkentin TE, Levine MN, Hirsh J, et al. Heparin-induced thrombocytopenia in patients treated with low-molecular-weight heparin or unfractionated heparin. N Engl J Med. 1995;332:1330–1335. doi:10.1056/NEJM199505183322003
32. Martel N, Lee J, Wells PS. Risk for heparin-induced thrombocytopenia with unfractionated and low-molecular-weight heparin thromboprophylaxis: a meta-analysis. Blood. 2005;106:2710–2715. doi:10.1182/blood-2005-04-1546
33. Warkentin TE, Sheppard JA, Horsewood P, et al. Impact of the patient population on the risk for heparin-induced thrombocytopenia. Blood. 2000;96:1703–1708.
34. Warkentin TE, Kelton JG. A 14-year study of heparin-induced thrombocytopenia. Am J Med. 1996;101:502–507. doi:10.1016/s0002-9343(96)00258-6
35. Warkentin TE, Kelton JG. Temporal aspects of heparin-induced thrombocytopenia. N Engl J Med. 2001;344:1286–1292. doi:10.1056/NEJM200104263441704
36. Lo GK, Juhl D, Warkentin TE, et al. Evaluation of pretest clinical score (4 T's) for the diagnosis of heparin-induced thrombocytopenia in two clinical settings. J Thromb Haemost. 2006;4:759–765. doi:10.1111/j.1538-7836.2006.01787.x
37. Cuker A, Gimotty PA, Crowther MA, Warkentin TE. Predictive value of the 4Ts scoring system for heparin-induced thrombocytopenia: a systematic review and meta-analysis. Blood. 2012;120:4160–4167. doi:10.1182/blood-2012-07-443051
38. Warkentin TE, Elavathil LJ, Hayward CP, et al. The pathogenesis of venous limb gangrene associated with heparin-induced thrombocytopenia. Ann Intern Med. 1997;127:804–812. doi:10.7326/0003-4819-127-9-199711010-00005
39. Lewis BE, Wallis DE, Berkowitz SD, et al. Argatroban anticoagulant therapy in patients with heparin-induced thrombocytopenia. Circulation. 2001;103:1838–1843. doi:10.1161/01.cir.103.14.1838
40. Warkentin TE, Pai M, Linkins LA. Direct oral anticoagulants for treatment of HIT: update of Hamilton experience and literature review. Blood. 2017;130:1104–1113. doi:10.1182/blood-2017-04-778993
41. Dyke CM, Smedira NG, Koster A, et al. A comparison of bivalirudin to heparin with protamine reversal in patients undergoing cardiac surgery with cardiopulmonary bypass: the EVOLUTION-ON study. J Thorac Cardiovasc Surg. 2006;131:533–539. doi:10.1016/j.jtcvs.2005.09.057
42. Warkentin TE, Greinacher A. Heparin-induced thrombocytopenia and cardiac surgery. Ann Thorac Surg. 2003;76:2121–2131. doi:10.1016/j.athoracsur.2003.09.034
43. Greinacher A, Selleng K, Warkentin TE. Autoimmune heparin-induced thrombocytopenia. J Thromb Haemost. 2017;15:2099–2114. doi:10.1111/jth.13813
44. Greinacher A, Thiele T, Warkentin TE, et al. Thrombotic thrombocytopenia after ChAdOx1 nCov-19 vaccination. N Engl J Med. 2021;384:2092–2101. doi:10.1056/NEJMoa2104840
45. Byrne RA, Rossello X, Coughlan JJ, et al. 2023 ESC Guidelines for the management of acute coronary syndromes. Eur Heart J. 2023;44:3720–3826. doi:10.1093/eurheartj/ehad191
46. Warkentin TE, Sheppard JA, Sigouin CS, Kohlmann T, Eichler P, Greinacher A. Gender imbalance and risk factor interactions in heparin-induced thrombocytopenia. Blood. 2006;108:2937–2941. doi:10.1182/blood-2005-11-012450