Heparin-Induced Thrombocytopenia (HIT)
Heparin-Induced Thrombocytopenia (HIT)
References
- American Society of Hematology 2018 guidelines for management of venous thromboembolism: heparin-induced thrombocytopenia — Blood Advances, American Society of Hematology (Cuker A, Arepally GM, Chong BH, et al.). Blood Adv 2018;2(22):3360-3392
- Evaluation of pretest clinical score (4 T's) for the diagnosis of heparin-induced thrombocytopenia in two clinical settings — Journal of Thrombosis and Haemostasis (Lo GK, Juhl D, Warkentin TE, et al.). J Thromb Haemost 2006;4(4):759-765
- Practical guide to the diagnosis and management of heparin-induced thrombocytopenia — Hematology, American Society of Hematology Education Program (May J, Cuker A). Hematology Am Soc Hematol Educ Program 2024;2024(1):388-395
Clinical Snapshot
Heparin-induced thrombocytopenia (HIT) is an immune-mediated adverse drug reaction caused by antibodies against complexes of platelet factor 4 (PF4) and heparin. These antibodies bind Fc receptors on platelets, paradoxically activating them and the coagulation cascade rather than simply destroying them — the result is thrombocytopenia together with a markedly increased risk of new venous and arterial thrombosis. HIT is a thrombotic disorder, not primarily a bleeding disorder, despite the low platelet count, and management decisions must not wait for laboratory confirmation given how quickly thrombosis can develop.[1] Onset is typically day 5-10 of a first heparin exposure (typical-onset HIT), or within 24 hours if the patient was exposed to heparin in the preceding roughly 100 days (rapid-onset HIT, from pre-existing antibodies). HIT can occur with unfractionated heparin (higher risk) or with low-molecular-weight heparin (lower but real risk); fondaparinux essentially does not cause it.
Recognize the Condition
- Platelet count fall of over 50% from baseline, typically to a nadir rarely below 20 ×10⁹/L — a very low count is atypical for HIT and should prompt consideration of another cause.
- Timing: onset day 5-10 of a first heparin exposure without recent prior exposure; as early as day 1 with heparin exposure in the preceding roughly 100 days; delayed-onset HIT can rarely present days after heparin has already been stopped.
- New venous thromboembolism (DVT/PE) or arterial thrombosis (limb ischemia, stroke, MI) — this occurs in a substantial proportion of HIT patients and drives most of its morbidity and mortality.
- Skin necrosis at a heparin injection site.
- An acute systemic reaction (fever, chills, dyspnea, chest pain, hemodynamic instability) temporally associated with an IV heparin bolus — a distinctive, easily missed presentation.
- Bleeding is uncommon and is not the primary clinical concern despite the thrombocytopenia.
The 5 clinical phases of HIT
Suspected HIT
Thrombocytopenia or a compatible presentation with confirmatory testing pending — management proceeds on clinical suspicion.
Acute HIT
Thrombocytopenia present with a positive laboratory result (or strong clinical suspicion) — full-dose non-heparin anticoagulation required.
Subacute HIT A
Platelet count has recovered, but a functional (platelet activation) assay remains positive — continued heparin avoidance and non-heparin anticoagulation as indicated.
Subacute HIT B
Platelet count has recovered and the functional assay is now negative, though the immunoassay may still be positive — heparin avoidance continues but the acute thrombosis risk has substantially declined.
Remote HIT
History of HIT with all current assays negative — heparin re-exposure may be considered only in specific, unavoidable circumstances (for example, cardiac surgery requiring cardiopulmonary bypass) under specialist direction.
Risk Factors
- Type and duration of heparin exposure — unfractionated heparin carries higher risk than low-molecular-weight heparin, and therapeutic-dose exposure carries higher risk than prophylactic-dose; even flush-dose or catheter-lock heparin can rarely cause HIT.
- Surgical patients, especially cardiac and orthopedic surgery, are at higher risk than medical patients.
- Female sex (modestly higher risk).
- Heparin exposure within the preceding roughly 100 days — raises the risk of rapid-onset HIT on re-exposure.
- Duration of heparin exposure beyond 4 days, though onset can be earlier with prior exposure.
Red Flags & Refer
Any suspected HIT is itself the red flag — do not wait for laboratory confirmation before acting on intermediate or high clinical probability:
- New arterial or venous thrombosis in a patient with recent or current heparin exposure and a falling platelet count — treat as HIT with thrombosis until proven otherwise; this carries the highest morbidity and mortality.
- An acute systemic or anaphylactoid-type reaction temporally linked to an IV heparin bolus.
- Skin necrosis at a heparin injection site.
- A severe or rapidly falling platelet count in any patient on heparin, regardless of the original reason heparin was started.
Assessment & Diagnosis
Differentiating the causes of new thrombocytopenia
| Parameter | Primary ITP | Heparin-induced thrombocytopenia | TTP / HUS | DIC | Pseudothrombocytopenia |
|---|---|---|---|---|---|
| Context | Isolated finding in an otherwise well patient; may follow a viral illness | Heparin exposure in the preceding 2 weeks, or within 24 hours if re-exposed within the preceding 100 days | Acutely unwell with neurologic changes, renal dysfunction, or fever | Sepsis, trauma, an obstetric emergency, or malignancy — a clinically obvious trigger | Incidental lab finding in an otherwise well patient |
| Onset | Variable, often gradual | Day 5-10 of a first heparin exposure; within 24 hours on re-exposure | Acute, rapidly progressive | Acute, tracking the underlying trigger | None — a sample-handling artifact |
| Platelet count pattern | Isolated thrombocytopenia; other cell lines normal | A fall of over 50% from baseline, rarely below 20 ×10⁹/L | Severe thrombocytopenia with microangiopathic hemolysis | Falling platelets with a prolonged PT/aPTT and low fibrinogen | Falsely low count only — the true count is normal |
| Peripheral smear | Normal morphology; large platelets may be seen | Normal platelet morphology | Schistocytes (fragmented red cells) | Schistocytes may be present; evidence of consumption | Platelet clumps or platelet satellitism around neutrophils |
| Key distinguishing test | Diagnosis of exclusion — no single confirmatory test | 4Ts score plus an anti-PF4/heparin antibody immunoassay | Severely reduced ADAMTS13 activity | Prolonged PT/aPTT, low fibrinogen, elevated D-dimer | Repeat the count in a citrate (blue-top) tube — it normalizes |
| Predominant risk | Bleeding | Thrombosis — venous and arterial — despite the low platelet count | Microvascular thrombosis and organ ischemia | Both bleeding and thrombosis | None — not a true hematologic abnormality |
The 4Ts pretest probability score
The 4Ts score is a 4-item clinical scoring system used before ordering laboratory testing — each item is scored 0, 1, or 2 points, for a total of 0-8.[2]
| Category | 2 points | 1 point | 0 points |
|---|---|---|---|
| Thrombocytopenia | Platelet fall over 50%, and nadir at least 20 ×10⁹/L | Platelet fall 30-50%, or nadir 10-19 ×10⁹/L | Platelet fall under 30%, or nadir under 10 ×10⁹/L |
| Timing of platelet fall | Days 5-10, or 1 day if prior heparin exposure within 30 days | Onset after day 10, or unclear timing, or 1 day if prior exposure 30-100 days ago | Platelet fall under 4 days without recent prior exposure |
| Thrombosis or other sequelae | New confirmed thrombosis, skin necrosis at an injection site, or an acute systemic reaction after IV heparin | Progressive or recurrent thrombosis, a non-necrotizing skin lesion, or suspected (unconfirmed) thrombosis | None |
| oTher causes excluded | No other apparent cause for the thrombocytopenia | A possible other cause is present | A definite other cause is present |
Total score interpretation: 0-3 points is low probability — HIT can effectively be excluded, and antibody testing is not indicated. 4-5 points is intermediate probability, and 6-8 points is high probability — both warrant immediate empiric non-heparin anticoagulation and antibody testing.[2] This is a genuine simple point-sum in structure, but it is scored manually here rather than through a live calculator — score each of the 4 categories against the table above and sum the total.
Confirmatory laboratory testing
An anti-PF4/heparin immunoassay (ELISA) is highly sensitive but only moderately specific — many weakly positive results are not clinically significant. A functional assay (serotonin release assay or heparin-induced platelet activation assay) is more specific and used to confirm an intermediate/high-probability or equivocal immunoassay result, though it is typically a reference-laboratory test that can take days to return.[3] Do not order antibody testing in a patient with a low 4Ts score — in this group testing mainly generates false positives and unnecessary anticoagulant changes.[1] Treatment decisions in an intermediate/high-probability patient must not wait for either assay's result.
Treatment Targets
- Stop all heparin exposure immediately.
- Prevent new or progressive thrombosis with an effective non-heparin anticoagulant.
- Avoid unnecessary bleeding from over-anticoagulation.
- Avoid warfarin-induced skin necrosis/venous limb gangrene by not starting warfarin until the platelet count has recovered.
- Confirm or exclude the diagnosis definitively and achieve platelet count recovery.
- Document HIT clearly for lifelong heparin avoidance.
Choose Initial Treatment
Intermediate or high 4Ts probability — stop all heparin immediately (including flushes and heparin-coated catheters) and start a non-heparin anticoagulant at a therapeutic (not prophylactic) dose right away, even before laboratory confirmation — this applies to isolated HIT (no imaging-confirmed thrombosis) as well, since the subsequent thrombosis risk is substantial.[1]
Choice of non-heparin anticoagulant — argatroban (hepatically cleared) is preferred with significant renal impairment or dialysis; bivalirudin is often preferred for cardiac surgery/PCI or combined renal-hepatic concerns given its short half-life; fondaparinux is a reasonable alternative in a hemodynamically stable patient with preserved renal function (avoid if ClCr under 30 mL/min); danaparoid is another parenteral alternative. Direct oral anticoagulants (apixaban, rivaroxaban) are increasingly used once the patient is clinically stable, as an alternative to a continued parenteral agent for the subacute phase.[3]
Do not start warfarin until the platelet count has recovered to at least 150 ×10⁹/L and the patient is stable on a non-heparin anticoagulant; overlap the 2 agents for a minimum of 5 days with confirmed platelet recovery before stopping the parenteral agent — starting warfarin while platelets remain low and unopposed by adequate parenteral anticoagulation risks warfarin-induced skin necrosis or venous limb gangrene.[1]
Duration of anticoagulation — treat isolated HIT (no thrombosis) for a minimum of about 4 weeks; treat HIT with thrombosis (HITT) for the same duration as any other VTE, typically at least 3 months.
Future heparin avoidance — document HIT clearly in the record and avoid all heparin products (unfractionated and low-molecular-weight) for life where possible, including for future surgery, dialysis, or cardiac catheterization, unless a hematology-directed protocol determines heparin is truly unavoidable after antibody titers have declined.
Non-Pharmacologic Treatment
- Discontinue all heparin sources immediately — IV infusions, subcutaneous prophylactic doses, heparin flushes, and heparin-coated catheters.
- Do not simply switch from unfractionated heparin to low-molecular-weight heparin (or vice versa) as the response to suspected HIT — cross-reactivity is common and this does not solve the problem.
- Avoid platelet transfusion unless there is active severe bleeding or an urgent invasive procedure requires correction — transfusion carries a theoretical risk of fueling further thrombosis in HIT.
- Use mechanical VTE prophylaxis (compression devices) as an adjunct while pharmacologic anticoagulation is being established.
- Advise the patient to wear medical alert identification noting the HIT history and to inform every future healthcare provider before any procedure that might involve heparin.
Pharmacologic Treatment Options
| Drug / intervention | Role | Dose / use | Important adverse effects | Teaching points |
|---|---|---|---|---|
| Argatroban Preferred | Non-heparin anticoagulant, especially renal impairment or dialysis | 2 mcg/kg/min IV infusion (reduce to 0.5-1.2 mcg/kg/min in hepatic impairment or critical illness), titrate to aPTT 1.5-3x baseline | Bleeding | Hepatically cleared — no renal dose adjustment needed. Prolongs the INR, complicating a warfarin transition — use a chromogenic factor X assay instead. |
| Bivalirudin Preferred | Especially cardiac surgery/PCI, or combined renal-hepatic concerns | 0.15-0.2 mg/kg/hour IV infusion, titrate to aPTT 1.5-2.5x baseline (institutional PCI protocols differ) | Bleeding | Short half-life allows rapid offset if bleeding occurs. Dose-reduce in renal impairment. |
| Fondaparinux Alternative | Hemodynamically stable patients with preserved renal function | 5-10 mg SC once daily, weight-based (5 mg under 50 kg, 7.5 mg 50-100 kg, 10 mg over 100 kg) | Bleeding | Avoid if ClCr under 30 mL/min. Does not require aPTT monitoring. Does not cross-react with HIT antibodies. |
| Danaparoid Alternative | Heparinoid with low clinical cross-reactivity | Weight-based loading dose then continuous infusion per institutional protocol | Bleeding | Some in-vitro structural cross-reactivity with HIT antibodies exists, but clinical cross-reactivity is low. Monitor anti-Xa levels where available — aPTT is not reliable. |
| Warfarin Special situation | Long-term anticoagulation only after platelet recovery and non-heparin overlap | Start a low dose only once the platelet count is at least 150 ×10⁹/L and the patient is stable on a non-heparin agent; overlap a minimum of 5 days | Bleeding, warfarin-induced skin necrosis if started prematurely | Never start as monotherapy in acute HIT. Use a chromogenic factor X assay, not INR, to guide the transition from argatroban. |
| Apixaban Alternative | Once clinically stable, for ongoing/subacute treatment | Per standard VTE treatment dosing for the selected agent | Bleeding | Increasingly used once the acute phase has passed and the platelet count is recovering; check the Drug Database entry for exact dosing. |
Special Populations
Pregnancy
HIT is rare in pregnancy — low-molecular-weight heparin, the preferred anticoagulant in pregnancy generally, carries lower HIT risk than unfractionated heparin. If HIT occurs in pregnancy, danaparoid has historically been preferred (extensive pregnancy-safety experience, does not cross the placenta); fondaparinux is an alternative with more limited but growing pregnancy data. Argatroban or bivalirudin are used when parenteral direct thrombin inhibition is specifically needed, under specialist direction. Warfarin is contraindicated in pregnancy (teratogenic) regardless of HIT status, and DOACs are not used in pregnancy.
Breastfeeding
Warfarin is compatible with breastfeeding once otherwise appropriate for the clinical situation. Data on argatroban, bivalirudin, danaparoid, and fondaparinux in breastfeeding are limited — a specialist-directed, individualized risk-benefit decision.
Pediatrics
HIT is uncommon in children outside settings with intensive heparin exposure (cardiac surgery, ECMO). When it occurs, non-heparin anticoagulant dosing is weight-based and specialist-directed — both argatroban and bivalirudin have published pediatric dosing, but treatment is always individualized.
Renal impairment
Argatroban is preferred (hepatic clearance, no renal adjustment needed). Avoid fondaparinux if ClCr is under 30 mL/min. Danaparoid is partially renally cleared — use caution or consider an alternative in severe renal impairment.
Hepatic impairment
Argatroban requires a significant dose reduction given impaired clearance. Bivalirudin is often preferred here given its more balanced (enzymatic plus partial renal) clearance.
Monitoring & Follow-Up
Daily platelet counts until recovery, then per standard follow-up. Monitor aPTT (argatroban, bivalirudin, or danaparoid without anti-Xa availability) or anti-Xa levels (fondaparinux, or danaparoid where available), titrated per institutional protocol for the agent chosen. Monitor clinically for new or progressive thrombosis throughout the acute phase. Use a chromogenic factor X assay — not INR — during the argatroban-to-warfarin transition, since argatroban falsely elevates the INR. Confirm platelet recovery to at least 150 ×10⁹/L before or while overlapping warfarin, and continue the overlap for a minimum of 5 days with 2 consecutive therapeutic INRs before stopping the parenteral agent. Document HIT clearly for future heparin-avoidance decisions — routine trending of anti-PF4 antibody titers is not needed for ongoing management, since they wane over weeks to months. Monitor limb perfusion and skin integrity given both the thrombosis risk and the necrosis risk if warfarin is started prematurely.
Treatment Failure / Next Step
- Confirm heparin has genuinely been stopped from every source, including flushes and heparin-coated catheters — ongoing occult heparin exposure is a common cause of apparent treatment failure.
- Confirm the non-heparin anticoagulant is dosed adequately and that monitoring (aPTT or anti-Xa) is in the therapeutic range, adjusted for renal/hepatic status.
- New or progressive thrombosis despite adequate non-heparin anticoagulation: escalate care (imaging, consider mechanical/interventional therapy for limb-threatening arterial thrombosis) and involve hematology urgently.
- Platelet count not recovering as expected: reconsider the diagnosis (was the 4Ts score and serology genuinely consistent with HIT, or is another cause of thrombocytopenia present) and consider hematology consultation or bone marrow evaluation if truly refractory.
- If warfarin-associated skin lesions or limb ischemia develop after a premature warfarin start: stop warfarin immediately, reverse with vitamin K, and resume or continue therapeutic non-heparin parenteral anticoagulation.
Common Errors to Avoid
- Waiting for laboratory confirmation before stopping heparin and starting a non-heparin anticoagulant in a patient with intermediate or high clinical probability.
- Switching from unfractionated heparin to low-molecular-weight heparin (or vice versa) as the response to suspected HIT.
- Giving a platelet transfusion for an incidentally low platelet count in suspected HIT without a clear bleeding indication.
- Starting warfarin before the platelet count has recovered and before adequate overlap with a non-heparin anticoagulant — risking warfarin-induced skin necrosis or venous limb gangrene.
- Using the INR, rather than a chromogenic factor X assay, to guide the argatroban-to-warfarin transition, leading to premature or inappropriate warfarin dosing.
- Ordering HIT antibody testing in a patient with a low 4Ts score, generating a false-positive result and an unnecessary anticoagulant switch.
- Treating isolated HIT (no imaging-confirmed thrombosis) as lower risk and using only prophylactic-dose anticoagulation — therapeutic dosing is needed given the high subsequent thrombosis risk.
- Failing to document HIT clearly in the patient's record, risking inadvertent heparin re-exposure during a future hospitalization or procedure.
Clinical Decision Algorithm
Drug Tables
How to use this table: doses are adult doses and require patient-specific confirmation, including renal/hepatic adjustment. Select any drug name to open its complete Pharmasist Drug Database page.
Non-heparin parenteral anticoagulants
Start immediately on intermediate/high clinical suspicion — do not wait for laboratory confirmation.
| Drug | Dosage | Adverse effects | Drug interactions | Comments |
|---|---|---|---|---|
| Argatroban | Standard infusion: 2 mcg/kg/min IV, titrate to aPTT 1.5-3x baseline Hepatic impairment: Reduce to 0.5-1.2 mcg/kg/min | Bleeding | Prolongs the INR, complicating a warfarin transition — use a chromogenic factor X assay instead | Hepatically cleared — no renal dose adjustment needed. Preferred with significant renal impairment or dialysis. |
| Bivalirudin | Infusion: 0.15-0.2 mg/kg/hour IV, titrate to aPTT 1.5-2.5x baseline (PCI protocols differ) | Bleeding | None significant | Short half-life allows rapid offset. Dose-reduce in renal impairment. Often preferred for cardiac surgery/PCI. |
| Fondaparinux | Under 50 kg: 5 mg SC once daily 50-100 kg: 7.5 mg SC once daily Over 100 kg: 10 mg SC once daily Renal impairment: Avoid if ClCr under 30 mL/min | Bleeding | None significant | Does not cross-react with HIT antibodies. Does not require aPTT monitoring. |
| Danaparoid | Dosing: Weight-based loading dose then continuous infusion, per institutional protocol | Bleeding | None significant | Some in-vitro structural cross-reactivity with HIT antibodies exists, but clinical cross-reactivity is low. Monitor anti-Xa levels where available. |
Transition and oral anticoagulation (subacute phase)
Warfarin must never be started as monotherapy in acute HIT.
| Drug | Dosage | Adverse effects | Drug interactions | Comments |
|---|---|---|---|---|
| Warfarin | Adults: Individualized low starting dose, titrated to INR | Bleeding, warfarin-induced skin necrosis if started prematurely | Numerous — review the full interaction profile before starting | Start only once the platelet count is at least 150 ×10⁹/L and the patient is stable on a non-heparin agent; overlap a minimum of 5 days with 2 consecutive therapeutic INRs before stopping the parenteral agent. Use a chromogenic factor X assay, not INR, during an argatroban overlap. |
| Apixaban | Adults: Per standard VTE treatment dosing | Bleeding | Strong dual CYP3A4/P-glycoprotein inhibitors or inducers significantly affect levels | Increasingly used once the acute phase has passed and the platelet count is recovering. |
| Rivaroxaban | Adults: Per standard VTE treatment dosing | Bleeding | Strong dual CYP3A4/P-glycoprotein inhibitors or inducers significantly affect levels | An alternative DOAC for the subacute/stable phase. |
| Dabigatran etexilate | Adults: Per standard VTE treatment dosing | Bleeding | P-glycoprotein inhibitors/inducers significantly affect levels | Requires a parenteral lead-in per standard VTE dosing, unlike apixaban/rivaroxaban. |
Agents to avoid in suspected or confirmed HIT
| Agent | Why avoided |
|---|---|
| Heparin (unfractionated) | The causative agent — stop immediately, including flushes and heparin-coated catheters. |
| Any low-molecular-weight heparin (for example, enoxaparin) | Substantial cross-reactivity with HIT antibodies — switching from UFH to LMWH (or vice versa) does not solve the problem. |
| Warfarin as monotherapy, or started before platelet recovery | Risks warfarin-induced skin necrosis and venous limb gangrene while the platelet count remains low and unopposed by adequate parenteral anticoagulation. |
Evidence note: drug inventory, dosing, adverse effects, and comments reflect standard Canadian practice, cross-checked against current ASH guidance.[1]
Professional reference only. Apply clinical judgment and verify patient-specific decisions against current authoritative guidance and product information.