Residency · Residency · Hematology Thrombosis

Heparin-Induced Thrombocytopenia

Introduction

Heparin-induced thrombocytopenia (HIT) is an immune-mediated, prothrombotic disorder that arises from the development of antibodies against complexes of platelet factor 4 (PF4) and heparin. It represents one of the most clinically consequential drug-related adverse effects in medicine, affecting 0.5 to 5% of heparin-exposed patients, with the highest incidence observed in surgical patients receiving unfractionated heparin (3 to 5%) and a substantially lower incidence with low-molecular-weight heparin (approximately 0.5%). The central paradox of HIT is that it produces thrombocytopenia through platelet consumption yet manifests primarily with thrombotic rather than hemorrhagic complications. Without prompt recognition and appropriate treatment, HIT carries a mortality rate of 10 to 20%, driven almost entirely by the devastating thrombotic sequelae that include deep vein thrombosis, pulmonary embolism, limb-threatening arterial occlusion, stroke, and organ infarction.

Pathophysiology

Type 1 HIT (Non-Immune)

Type 1 HIT is a non-immune phenomenon resulting from direct heparin-platelet interaction that causes mild platelet aggregation. The platelet count drop is modest, rarely falling below 100,000, and occurs within the first 2 days of heparin exposure. This entity has no clinical significance, resolves spontaneously without intervention, and does not require heparin discontinuation. It is not considered true HIT and should be clearly distinguished from the immune-mediated type 2 disorder, which carries entirely different clinical implications.

Type 2 HIT (Immune-Mediated) - "HIT"

The pathogenesis of type 2 HIT, which is the clinically relevant entity referred to simply as "HIT" throughout clinical practice, proceeds through a sequence of immunologic steps that ultimately generate a profound prothrombotic state. In the first step, heparin molecules bind to platelet factor 4 (PF4), a cationic chemokine stored in platelet alpha granules and released upon platelet activation. This binding induces a conformational change in PF4 that exposes a neoepitope not present on native PF4 alone. In the second step, the immune system generates IgG antibodies directed against this PF4/heparin neoepitope. In the third step, these IgG antibodies form immune complexes on the platelet surface, and the Fc portion of the IgG molecules engages Fc-gamma-RIIa receptors on adjacent platelets, triggering intense platelet activation with the release of procoagulant microparticles, thromboxane A2 generation, and further PF4 release that amplifies the cycle. In the fourth step, the massively activated platelets are consumed within forming thrombi throughout the vasculature, producing the characteristic combination of thrombocytopenia and thrombosis.

The prothrombotic cascade extends beyond platelets. The pathogenic antibodies also activate endothelial cells and monocytes, inducing tissue factor expression on their surfaces and further amplifying coagulation activation. This multi-cellular activation explains the intensity of the thrombotic response in HIT. An important feature of HIT immunology is that the antibodies are transient, typically declining to undetectable levels over 3 to 4 months. Unlike most autoimmune conditions, HIT antibodies do not persist indefinitely, a characteristic that has implications for the re-exposure of patients with a remote history of HIT to heparin.

Risk Factors

The risk of developing HIT varies substantially with the type of heparin, the duration of exposure, and the clinical setting. Unfractionated heparin carries a far greater risk than LMWH, which in turn carries a greater risk than fondaparinux, the latter having essentially zero risk because its pentasaccharide structure does not bind PF4. Exposure to heparin for more than 4 days increases risk as the immune response requires time to develop. The clinical setting strongly influences risk, with surgical patients, particularly those undergoing orthopedic and cardiac surgery, at higher risk than medical patients. Female sex may confer a modestly increased risk.

Clinical Features

Timing (The "4 T's" - Timing)

The temporal relationship between heparin exposure and the development of thrombocytopenia is one of the most diagnostically useful features of HIT. In the typical presentation, the platelet count begins to fall between day 5 and day 14 of heparin exposure, reflecting the time required for the primary immune response to generate pathogenic IgG antibodies. Rapid-onset HIT occurs within 24 hours of heparin re-exposure in patients who have circulating HIT antibodies from a recent prior heparin exposure (generally within the preceding 100 days). This accelerated presentation reflects an anamnestic immune response rather than a primary immunization. Delayed-onset HIT is an important and often underrecognized variant in which thrombocytopenia and thrombosis develop days to weeks after heparin has been discontinued. This autoimmune-like variant involves antibodies capable of activating platelets in the absence of exogenous heparin and may present to clinicians who are unaware of the patient's prior heparin exposure.

Thrombocytopenia

The thrombocytopenia of HIT is typically moderate, with platelet count nadirs generally between 20,000 and 100,000. The most common pattern is a decline of 50% or more from the baseline platelet count, and it is critical to recognize that this relative decline may be diagnostically significant even when the absolute platelet count remains within the normal range. For example, a decline from 400,000 to 180,000 represents a greater than 50% drop and should raise suspicion for HIT despite the seemingly "normal" absolute value. HIT rarely produces severe thrombocytopenia below 15,000; when the platelet count drops to below 10,000, alternative diagnoses such as DIC, drug-induced immune thrombocytopenia from other medications, or sepsis-related thrombocytopenia should be strongly considered.

Thrombosis

Thrombotic complications occur in 30 to 50% of patients with HIT who are not treated with non-heparin anticoagulation, underscoring the critical importance of therapeutic anticoagulation even in patients who have not yet manifested clinically apparent thrombosis. Venous thrombosis accounts for approximately 60% of HIT-associated thrombotic events and includes DVT and PE as the most common manifestations, along with less common but important entities such as adrenal vein thrombosis (which can produce bilateral hemorrhagic adrenal infarction and acute adrenal insufficiency), cerebral venous sinus thrombosis, and hepatic or portal vein thrombosis. Arterial thrombosis accounts for the remaining 40% and may present as limb artery occlusion (historically termed "white clot syndrome" due to the platelet-rich composition of HIT thrombi), stroke, myocardial infarction, or mesenteric ischemia.

Skin necrosis can occur at subcutaneous heparin injection sites or at distant locations and reflects dermal microvascular thrombosis. Venous limb gangrene is a devastating complication that occurs when warfarin is initiated during the acute phase of HIT without adequate non-heparin anticoagulation. The mechanism involves the rapid depletion of protein C (which has a half-life of approximately 8 hours) relative to the procoagulant factors, creating a transient hypercoagulable state superimposed on the already activated HIT-related thrombotic process. This complication is entirely preventable by adhering to the principle that warfarin must never be started until the platelet count has recovered to above 150,000 and therapeutic non-heparin anticoagulation is established.

<image>A pathophysiology diagram of heparin-induced thrombocytopenia type 2. Show the step-by-step immunologic mechanism: (1) Heparin molecules binding to PF4 on platelet surface, causing conformational change exposing neoepitope. (2) IgG antibody production against PF4/heparin complex. (3) IgG Fc portion binding to FcγRIIa receptor on adjacent platelets, causing intense platelet activation with microparticle release and thromboxane A2 generation. (4) Massive thrombin generation from activated platelets and monocytes (tissue factor expression). (5) Platelet consumption in forming thrombi leading to thrombocytopenia. Show the clinical consequences: venous thrombosis (DVT, PE), arterial thrombosis (limb ischemia, stroke), skin necrosis, and adrenal infarction. Include a timeline showing typical onset (day 5-14), rapid onset (<24h if recent exposure), and delayed onset (post-heparin discontinuation). Medical immunology illustration style with molecular detail and clinical correlation.</image>

Diagnosis

4T's Score (Pre-Test Probability)

The 4T's scoring system is the validated clinical tool for estimating the pretest probability of HIT and guiding the decision to pursue further laboratory testing. It evaluates four domains, each scored from 0 to 2 points. Thrombocytopenia is scored based on the magnitude of platelet count decline and the nadir (a fall of 50% or more with a nadir of 20,000 or above scores 2 points; a fall of 30 to 50% or nadir of 10,000 to 19,000 scores 1 point; a fall of less than 30% or nadir below 10,000 scores 0). Timing is assessed based on the onset of platelet decline relative to heparin exposure (day 5 to 10, or within 1 day if heparin exposure within the past 30 days scores 2 points; consistent timing but unclear, or onset after day 10 scores 1 point; onset within 4 days without recent prior exposure scores 0). Thrombosis is evaluated by the presence and type of thrombotic events (new confirmed thrombosis, skin necrosis, or acute systemic reaction after heparin bolus scores 2; progressive or recurrent thrombosis or suspected thrombosis scores 1; none scores 0). Other causes of thrombocytopenia are assessed (no other apparent cause scores 2; possible alternative cause scores 1; definite alternative cause scores 0).

A total score of 0 to 3 indicates low probability, with an approximately 5% likelihood of HIT and a negative predictive value exceeding 99%. In these patients, HIT is effectively excluded and further testing is not necessary. A score of 4 to 5 indicates intermediate probability, with a 14 to 25% likelihood of HIT, and should prompt both confirmatory laboratory testing and consideration of empiric non-heparin anticoagulation. A score of 6 to 8 indicates high probability, with a 60 to 80% likelihood of HIT, and mandates immediate cessation of all heparin and initiation of non-heparin anticoagulation while awaiting laboratory confirmation.

Laboratory Testing

The immunoassay for anti-PF4/heparin antibodies (ELISA) serves as the initial laboratory test. It offers excellent sensitivity exceeding 99% but only moderate specificity of 50 to 80% because it detects both pathogenic IgG antibodies and clinically insignificant IgA and IgM antibodies that do not cause platelet activation. The high negative predictive value of the ELISA means that a negative result effectively excludes HIT. Importantly, the optical density (OD) of the ELISA result correlates with the probability of clinical HIT: an OD value above 2.0 is strongly suggestive of pathogenic HIT antibodies, while an OD below 0.4 essentially excludes the diagnosis.

The serotonin release assay (SRA) is the gold standard confirmatory test, with sensitivity of approximately 95% and specificity of approximately 97%. It is a functional assay that measures heparin-dependent platelet activation by quantifying the release of radiolabeled 14C-serotonin from donor platelets in the presence of patient serum and varying concentrations of heparin. A serotonin release exceeding 20% at therapeutic heparin concentrations constitutes a positive result. The heparin-induced platelet aggregation (HIPA) assay is an alternative functional test available at some centers. The main limitation of functional assays is their restricted availability at reference laboratories, with turnaround times typically ranging from 2 to 5 days. The recommended diagnostic algorithm therefore follows a sequential approach: 4T's score for pretest probability, followed by ELISA for intermediate and high-probability cases, with SRA confirmation for positive ELISA results.

Approach

The clinical approach is driven by the 4T's score. For patients with a low score (0 to 3), no testing is needed and heparin can be continued. For patients with an intermediate or high score (4 to 8), all heparin must be immediately discontinued, including heparin line flushes, heparin-coated devices, and heparin locks. Non-heparin anticoagulation at therapeutic doses should be started simultaneously, and an ELISA sent. If the ELISA returns negative, HIT is unlikely and discontinuation of the non-heparin anticoagulant can be considered, though clinical judgment must be applied. If the ELISA is positive, particularly with a high OD value above 2.0, or if the SRA is positive, HIT is confirmed and treatment should continue as outlined below.

Treatment

Immediate Actions

The management of confirmed or strongly suspected HIT involves four critical actions. First, all sources of heparin must be eliminated, including not only therapeutic heparin infusions and subcutaneous injections but also heparin flushes for intravenous lines, heparin-coated central venous catheters, and heparin locks. Second, non-heparin anticoagulation must be initiated at therapeutic doses even in the absence of documented thrombosis, because the risk of developing thrombosis in untreated HIT patients exceeds 50%. Third, platelet transfusion must be avoided unless there is life-threatening hemorrhage or an essential surgical indication, as transfused platelets can be activated by the circulating HIT antibodies and may worsen thrombosis. Fourth, warfarin must not be started until the platelet count has recovered to above 150,000 and the patient is therapeutically anticoagulated with a non-heparin agent, to prevent the catastrophic complication of venous limb gangrene.

Non-Heparin Anticoagulants

Argatroban

Argatroban is a direct thrombin inhibitor administered as a continuous intravenous infusion, making it the most commonly used parenteral anticoagulant for acute HIT management. The starting dose is 2 mcg/kg/min, reduced to 0.5 mcg/kg/min in patients with hepatic impairment because argatroban is hepatically metabolized. Monitoring is performed with the aPTT, targeting 1.5 to 3 times the baseline value (typically 50 to 80 seconds), with levels checked every 2 hours until a stable therapeutic range is achieved.

A critical practical consideration with argatroban is that it falsely elevates the INR because it inhibits thrombin in the PT reagent system. This creates a significant challenge during the transition to warfarin, as the INR cannot be reliably used to determine when warfarin has achieved therapeutic anticoagulation. The recommended approach is to use a chromogenic factor X assay, with a target of 20 to 40%, to assess warfarin's anticoagulant effect independently of argatroban's interference. During the overlap period, argatroban and warfarin should be administered concurrently for at least 5 days. The argatroban infusion rate should be reduced to 2 mcg/kg/min (if higher doses have been used), and the INR should be rechecked 2 hours after temporarily holding the argatroban infusion. The half-life of argatroban is 39 to 51 minutes, which is advantageous in patients who may require procedures.

Bivalirudin

Bivalirudin is another direct thrombin inhibitor that is administered intravenously, with an even shorter half-life of approximately 25 minutes. Its primary role in HIT management is during cardiac catheterization and cardiac surgery, where its ultra-short half-life provides a favorable safety profile. For HIT, bivalirudin is dosed at 0.15 to 0.2 mg/kg/hr without a bolus and monitored with the aPTT or activated clotting time (ACT). Clearance is approximately 80% enzymatic and 20% renal, requiring only mild dose reduction in chronic kidney disease.

Fondaparinux (Off-Label for HIT)

Fondaparinux does not cross-react with HIT antibodies because it does not bind platelet factor 4. This pharmacologic property, combined with its predictable pharmacokinetics, once-daily subcutaneous administration, and absence of routine monitoring requirements, has made fondaparinux a widely used and practical option for HIT anticoagulation despite its off-label status in this indication. Its limitations include a long half-life of 17 to 21 hours (which can be problematic if urgent procedures are needed), absence of a specific reversal agent, and exclusive renal clearance that precludes use in severe renal impairment.

DOACs for HIT

Emerging evidence supports the use of rivaroxaban and apixaban for HIT, though their precise role in the acute management of HIT with active thrombosis is still being defined. Current practice generally involves initiating DOACs after an initial period of parenteral anticoagulation with a non-heparin agent and after platelet recovery has begun, though there is growing experience with direct DOAC initiation in acute HIT, particularly for patients without large-vessel thrombosis. DOACs are not yet endorsed by major guidelines as initial therapy for acute HIT with thrombosis.

Warfarin Transition

The transition to warfarin must follow strict criteria: warfarin can be introduced only after the platelet count has recovered to above 150,000 and the patient is receiving therapeutic non-heparin anticoagulation. The starting dose should not exceed 5 mg daily to minimize the risk of protein C depletion, and the overlap with the parenteral anticoagulant must be maintained for at least 5 days and until the INR has been at or above 2.0 for 24 hours. The required duration of anticoagulation following HIT depends on whether thrombosis occurred: HIT with documented thrombosis requires a minimum of 3 months of anticoagulation, while isolated HIT without thrombosis requires a minimum of 4 weeks, with some experts recommending 3 months.

Special Situations

HIT in Cardiac Surgery

Cardiac surgery presents unique challenges in HIT management because cardiopulmonary bypass traditionally requires heparin for anticoagulation. The rate of positive anti-PF4/heparin ELISA results following cardiac surgery is remarkably high, with up to 50% of post-CABG patients developing detectable antibodies. However, the vast majority of these antibodies are non-pathogenic, and clinical HIT develops in only a small fraction. For patients with acute HIT who require cardiac surgery, bivalirudin is the preferred intraoperative anticoagulant and can be used safely for cardiopulmonary bypass. For patients with remote HIT whose antibodies have become undetectable (typically more than 100 days after the episode), brief intraoperative heparin exposure for cardiopulmonary bypass is considered acceptable, provided non-heparin anticoagulation is used in the pre- and postoperative periods.

Autoimmune HIT (Delayed-Onset HIT)

Autoimmune HIT, also termed delayed-onset HIT, is a distinct variant that presents after heparin has been discontinued. The pathogenic antibodies in this variant have the unusual property of activating platelets at low heparin concentrations and even in the complete absence of heparin, as demonstrated by a positive SRA with buffer-only control (without added heparin). This autoimmune behavior makes heparin discontinuation alone insufficient, as the antibodies continue to drive platelet activation independently. Treatment involves non-heparin anticoagulation combined with intravenous immunoglobulin (IVIG) at 1 g/kg for 2 days, which may accelerate platelet recovery by blocking the Fc-gamma-RIIa receptor on platelets and inhibiting antibody-mediated activation. Plasma exchange has been reported for refractory cases.

<image>A diagnostic and treatment algorithm for HIT. Start with "Suspected HIT (platelet drop ≥50%, day 5-14 of heparin)." First step: calculate 4T's score. Score 0-3 (low): HIT unlikely → continue heparin, no testing needed. Score 4-5 (intermediate): stop all heparin → start non-heparin anticoagulation → send PF4/heparin ELISA. Score 6-8 (high): stop all heparin → start non-heparin anticoagulation (argatroban IV, bivalirudin IV, or fondaparinux SC) → send ELISA. ELISA result: if negative (OD <0.4) → HIT excluded → consider stopping non-heparin anticoagulant. If positive (especially OD >2.0) → confirm with SRA if available → confirmed HIT. Treatment continuation: non-heparin anticoagulation until platelets >150,000 → then transition to warfarin (start ≤5 mg daily, overlap ≥5 days + INR ≥2.0) or DOAC. Duration: HIT with thrombosis = 3 months minimum; isolated HIT = 4 weeks minimum. Include warning boxes: "NO platelet transfusion" and "NO warfarin until platelets >150K." Clean clinical algorithm with decision diamonds and action boxes.</image>

Key Clinical Pearls

• The 4T's score is essential: a score of 0-3 has a negative predictive value >99% for HIT; do not send PF4/heparin ELISA for low-probability cases
• ELISA OD values matter: OD >2.0 strongly predicts clinical HIT and functional assay positivity; OD <0.4 effectively excludes HIT
• Start non-heparin anticoagulation at THERAPEUTIC doses even without documented thrombosis; the risk of thrombosis is 30-50% in untreated HIT
• NEVER initiate warfarin during active HIT until platelets have recovered to >150,000; warfarin during acute HIT can cause venous limb gangrene
• Argatroban falsely elevates INR; use a chromogenic factor X assay (target 20-40%) to guide warfarin transition
• Fondaparinux is widely and effectively used off-label for HIT; its predictable pharmacokinetics, SC administration, and lack of cross-reactivity with HIT antibodies make it an attractive option
• HIT antibodies are transient (3-4 months); patients with remote HIT (>100 days, negative antibodies) can receive brief heparin exposure if needed (e.g., cardiac surgery)

References

1. Cuker A, et al. American Society of Hematology 2018 guidelines for management of venous thromboembolism: heparin-induced thrombocytopenia. Blood Adv. 2018;2(22):3360-3392.
2. Lo GK, 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(4):759-765.
3. Greinacher A. Heparin-Induced Thrombocytopenia. N Engl J Med. 2015;373(3):252-261.
4. Warkentin TE. Heparin-induced thrombocytopenia: pathogenesis and management. Br J Haematol. 2003;121(4):535-555.
5. Linkins LA, et al. Treatment and prevention of heparin-induced thrombocytopenia: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: ACCP Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2 Suppl):e495S-e530S.

Read this lecture as Markdown