HaematologyTier 1Disease (DEADMAN)

Heparin-induced thrombocytopenia

Description

An antibody-mediated PROTHROMBOTIC drug reaction that happens to lower the platelet count
- *The thrombocytopenia is the clue; thrombosis is the disease*
Thrombosis in up to 50% if heparin is simply stopped without alternative anticoagulation
Venous > arterial (~4:1) - DVT/PE, but also limb ischaemia, stroke, MI, adrenal haemorrhagic infarction, skin necrosis at injection sites

	Type I (non-immune)	Type II (immune) = "HIT"
Mechanism	Direct platelet aggregation	Anti-PF4/heparin IgG
Onset	1-4 days	5-10 days
Platelet nadir	>100, mild	~50-70, or a 50% fall
Thrombosis	No	Yes, high risk
Action	Continue heparin	Stop all heparin, anticoagulate with a non-heparin agent

Bleeding is distinctly unusual - a bleeding patient with thrombocytopenia on heparin usually has something else

Epidemiology

Unfractionated heparin ~10x the risk of LMWH
Surgical > medical > obstetric; highest after cardiopulmonary bypass (~1-3%)
Overall ~0.2-3% of heparin-exposed patients; ~0.1-1% with LMWH
F>M ~2:1
Rare in children, pregnancy and patients on fondaparinux
Very rare with fondaparinux, and rarer still with the DOACs

Aetiopathogenesis

Platelet factor 4 (PF4), a positively charged alpha-granule protein, binds negatively charged heparin
-> conformational change exposes a neoepitope
-> IgG anti-PF4/heparin antibodies
-> immune complexes bind FcgammaRIIa on platelets
- -> platelet activation, degranulation, microparticle release -> massive thrombin generation
-> complexes also activate monocytes and endothelium -> tissue factor expression
-> consumption of activated platelets = thrombocytopenia; thrombin excess = thrombosis

*The paradox resolved: platelets are being consumed by activation, not destroyed passively. Hence thrombosis, not bleeding.*
Chain length matters: UFH forms optimal PF4 complexes; LMWH less so; fondaparinux barely at all
Antibodies are transient - typically undetectable by 3 months (no anamnestic response, which is why re-exposure after 100 days can be considered)

Anti-PF4 disorders without heparin

VITT - after adenoviral-vector COVID-19 vaccines; anti-PF4 antibodies binding a heparin-independent site
- Very high D-dimer, low fibrinogen, severe thrombocytopenia, thrombosis at unusual sites - especially cerebral venous sinus and splanchnic
- Treat with IVIG + non-heparin anticoagulation; avoid platelet transfusion
Spontaneous/autoimmune HIT - after infection, surgery or knee replacement with no heparin exposure
Delayed-onset HIT - starts or worsens after heparin has stopped, up to 3 weeks later

Diagnosis

Timing - the strongest clue

Platelet fall begins day 5-10 of heparin exposure
Rapid-onset HIT (within hours) if heparin exposure in the preceding 30-100 days - pre-existing antibody
Delayed-onset HIT - up to 3 weeks after heparin has been stopped
*Check for heparin flushes, heparin-bonded lines and dialysis circuits - patients are often told they had "no heparin"*

4Ts score - do this before ordering any test

	2 points	1 point	0 points
Thrombocytopenia	Fall >50% AND nadir >=20	Fall 30-50%, or nadir 10-19	Fall <30% or nadir <10
Timing	Day 5-10, or <=1 day with heparin in the last 30 days	Consistent but unclear, or after day 10	<=4 days, no recent exposure
Thrombosis	New thrombosis, skin necrosis, or systemic reaction after IV bolus	Progressive/recurrent thrombosis, erythematous skin lesions, suspected thrombosis	None
oTher cause	None apparent	Possible	Definite

Total	Probability	Action
0-3 (low)	<5%	HIT effectively excluded - continue heparin, look elsewhere
4-5 (intermediate)	~10-20%	Stop heparin, start alternative, test
6-8 (high)	~50-80%	Stop heparin, start alternative, test

Laboratory

1. Immunoassay (anti-PF4/heparin ELISA or chemiluminescence)

Very high sensitivity (>99%), poor specificity - a negative result excludes HIT; a positive one does not confirm it
Optical density correlates with probability - OD >2.0 is far more likely to be true HIT than a weak positive
IgG-specific assays are preferred; polyspecific assays detect clinically irrelevant IgA/IgM

2. Functional assay (confirmatory)

Serotonin release assay (SRA) - the gold standard; or heparin-induced platelet activation (HIPA)
High specificity; slow, send-away, available in reference laboratories only
*Never wait for it before changing anticoagulation*

Imaging

Bilateral lower limb Doppler in every confirmed case, even without symptoms - asymptomatic DVT is common and changes duration of anticoagulation

Management

*Two things must happen simultaneously: stop heparin, and start a different anticoagulant. Stopping heparin alone is not treatment.*

A. Immediate

Stop ALL heparin - infusions, prophylactic doses, flushes, heparin-bonded catheters, dialysis circuits
Document as a drug allergy
Start a full-dose non-heparin anticoagulant immediately - the prothrombotic state persists for days to weeks

Agent	Notes
Argatroban	Direct thrombin inhibitor, IV, hepatically cleared - use in renal failure. Prolongs INR - complicates warfarin transition
Bivalirudin	Direct thrombin inhibitor; short half-life; used in cardiac surgery/PCI
Danaparoid	Heparinoid, anti-Xa monitoring; ~10% in vitro cross-reactivity
Fondaparinux	Convenient, SC once daily; renally cleared
DOACs (rivaroxaban, apixaban)	Now accepted for clinically stable HIT without severe thrombocytopenia, and for transition after platelet recovery

B. What NOT to do

*Do NOT start warfarin until platelets have recovered to >150*
- Warfarin depletes protein C faster than factors II/IX/X -> transient hypercoagulability
- -> venous limb gangrene and warfarin-induced skin necrosis
- If warfarin has already been started, reverse it with vitamin K and continue the parenteral agent
- Overlap with the non-heparin agent for >=5 days AND until INR therapeutic
*Do NOT transfuse platelets prophylactically - fuels thrombosis. Only for life-threatening bleeding*
*Do NOT substitute LMWH for UFH* - cross-reactivity is near-complete
Do not place an IVC filter

C. Duration of anticoagulation

HIT with thrombosis (HITT): at least 3 months
Isolated HIT without thrombosis: until platelet recovery, then at least 4 weeks - thrombotic risk persists after the count normalises

D. Future heparin exposure

Antibodies are transient and usually gone by ~100 days
Cardiac surgery requiring bypass: if anti-PF4 antibodies have cleared, a single intraoperative UFH exposure is acceptable, with a non-heparin agent before and after
- Otherwise use bivalirudin for bypass
Label the patient; heparin avoidance should be lifelong unless specifically re-assessed

E. VITT

IVIG 1 g/kg for 2 days + non-heparin anticoagulation
Corticosteroid, plasma exchange in refractory cases
*Avoid platelet transfusion and heparin*
Fibrinogen replacement if <1.5 g/L

Associations

Cardiopulmonary bypass and cardiac surgery (highest incidence)
Orthopaedic surgery, trauma, major vascular surgery
Haemodialysis (circuit heparin)
ECMO, ventricular assist devices, intra-aortic balloon pumps
VITT after adenoviral-vector COVID-19 vaccines (ChAdOx1, Ad26)
Spontaneous/autoimmune HIT - post-infection, post-arthroplasty
Warfarin-induced venous limb gangrene and skin necrosis
Adrenal haemorrhage from bilateral adrenal vein thrombosis (-> adrenal insufficiency, easily missed)
Antiphospholipid syndrome (a differential for thrombosis with thrombocytopenia)

Natural history & complications

Platelet recovery typically within 4-14 days of stopping heparin
Antibodies disappear at a median ~50-85 days; usually undetectable by 100 days
*Thrombotic risk persists for weeks after the count normalises* - the reason for continued anticoagulation

Outcomes

Thrombosis in ~50% if untreated (heparin simply stopped)
Amputation ~10%; mortality ~5-20% in untreated or late-recognised disease
Early recognition and alternative anticoagulation reduce new thrombosis by ~50%

Complications

DVT/PE, arterial thrombosis, limb gangrene, stroke, MI
Warfarin-induced venous limb gangrene (preventable, and a classic exam scenario)
Bilateral adrenal haemorrhagic infarction -> adrenal crisis
Skin necrosis at heparin injection sites; anaphylactoid reaction after IV heparin bolus
Bleeding from the alternative anticoagulant

Pitfalls

Missed heparin flushes and heparin-bonded lines
Ordering the immunoassay in low-probability patients -> false positives -> unnecessary and hazardous anticoagulant switching
Attributing thrombocytopenia after cardiac surgery to haemodilution and bypass alone (expect a nadir at day 1-3 with recovery by day 5-7 - a second fall after day 5 is HIT until proven otherwise)

6 more sections, plus exam facts

Premium unlocks every note across every specialty, and the full exam fact library behind it.

Get premium access