HIT (Heparin-Induced Thrombocytopenia)
Immune-mediated platelet activation by anti-PF4/heparin antibodies. Paradoxically a PRO-thrombotic state despite low platelets. The platelets drop, but the patient clots. Stop ALL heparin immediately and switch to a non-heparin anticoagulant.
🔍 Overview
4T Score (Pretest Probability) -likelihood of heparin-induced thrombocytopenia (HIT); guides whether to send anti-PF4 and stop heparin
| Criterion | 2 Points | 1 Point | 0 Points |
|---|---|---|---|
| Thrombocytopenia | Drop > 50% AND nadir ≥ 20K | Drop 30–50% OR nadir 10–19K | Drop < 30% OR nadir < 10K |
| Timing | Days 5–10 OR ≤ 1 day if prior heparin within 30 days | Days 5–10 (unclear) OR > day 10 | < day 4 (no prior exposure) |
| Thrombosis | New thrombosis, skin necrosis, or anaphylaxis post-heparin bolus | Progressive or recurrent thrombosis | None |
| Other causes | No other cause for thrombocytopenia | Possible other cause | Definite other cause |
0–3: low probability (< 5%) → HIT unlikely, no further testing. 4–5: intermediate → send PF4 antibody, start non-heparin anticoagulant. 6–8: high probability → treat as HIT while awaiting confirmatory tests.
Key Features
- Timing: platelet drop typically 5–10 days after heparin initiation (or within 24h if prior heparin exposure within 30 days -rapid onset HIT)
- Platelet drop: usually > 50% from baseline (not absolute count -a drop from 300K to 130K is HIT)
- Thrombosis in 50%: DVT/PE (most common), arterial (stroke, MI, limb ischemia), skin necrosis at heparin injection sites
- UFH >> LMWH for HIT risk (but LMWH can also cause it -cross-reactivity ~90%)
🚨 Management
Immediate Actions
Stop ALL heparin immediately -IV, SC, line flushes, heparin-coated catheters. Even trace amounts perpetuate the immune response.
| Drug | Dose | Notes |
|---|---|---|
| Argatroban 1ST LINE | 0.5–2 mcg/kg/min IV (no bolus). Monitor aPTT. | Hepatically cleared -preferred in renal failure. Falsely elevates INR (complicates warfarin bridging). Reduce dose in hepatic impairment, ICU, post-cardiac surgery. |
| Bivalirudin 1ST LINE | 0.15–0.2 mg/kg/hr IV. Monitor aPTT. | Short half-life (~25 min). Preferred for PCI and cardiac surgery settings. Partially renally cleared. |
| Fondaparinux ALTERNATIVE | Weight-based SC (same as VTE dosing) | Pentasaccharide -does NOT cross-react with HIT antibodies. Off-label for HIT but widely used. No monitoring needed. Renally cleared -avoid if CrCl < 30. |
Do NOT transfuse platelets in HIT unless life-threatening bleeding. Platelets are prothrombotic in HIT, they become activated by HIT antibodies → more thrombosis. Do NOT start warfarin until platelets recover to ≥ 150K, early warfarin in HIT can cause protein C depletion → warfarin-induced skin necrosis and venous limb gangrene. When transitioning: overlap DTI (argatroban/bivalirudin) with warfarin for ≥ 5 days AND until INR is therapeutic on two consecutive days.
🧪 Workup
Workup
- 4T score -pre-test probability. ≥ 6 = high probability → start non-heparin anticoagulation immediately while awaiting confirmatory testing. 4-5 = intermediate → test. ≤ 3 = low → HIT unlikely. Components: Thrombocytopenia (% fall + nadir), Timing (day 5-10), Thrombosis (new), oTher cause (none identified). [4T Score Validation, 2006]
- PF4/heparin ELISA (PF4 antibody) -screening test. High sensitivity (~97%), moderate specificity (~75%). Negative ELISA essentially rules out HIT. Positive → need confirmatory test. OD > 2.0 strongly predictive.
- Serotonin release assay (SRA) -confirmatory gold standard. High specificity (~95%). Takes 3-7 days to result. Functional assay -measures actual platelet activation.
- CBC trend -platelet nadir typically 5-10 days after heparin exposure. Classic: > 50% drop from baseline (e.g., 250K → 80K). Nadir usually 20-80K. If < 20K → consider other diagnoses (DIC, TTP).
- Bilateral lower extremity duplex US -30-50% of HIT patients have occult DVT at diagnosis even without symptoms. Screen all confirmed/suspected HIT patients.
- Review ALL heparin exposure -IV drips, SQ prophylaxis, line flushes, heparin-coated catheters (dialysis, PICC lines), heparin in OR tubing. Even brief exposure counts.
- Timing clues: Typical onset day 5-10. Rapid-onset HIT (< 24h) = prior heparin exposure within 100 days (pre-formed antibodies). Delayed-onset HIT = develops after heparin stopped (rare, up to 3 weeks).
💊 Medications
Medications
| Drug | Dose | Route | Notes |
|---|---|---|---|
| Argatroban | 2 mcg/kg/min (0.5-1.2 in liver disease) | IV drip | Direct thrombin inhibitor. Argatroban HIT Trial, 2001 Hepatically metabolized -reduce dose in liver failure. Titrate to aPTT 1.5-3× baseline. Falsely elevates INR → complicates warfarin transition. |
| Bivalirudin | 0.15-0.25 mg/kg/hr | IV drip | Alternative DTI. Shorter half-life (25 min vs 45 min) -preferred for PCI, cardiac surgery, or renal failure. Does not elevate INR. |
| Fondaparinux | 5-10 mg SQ daily | SQ | Factor Xa inhibitor. Off-label for HIT but widely used. No IV monitoring needed. Minimal cross-reactivity with HIT antibodies (< 1%). Renally cleared -avoid if CrCl < 30. |
| DO NOT use LMWH | - | - | 90% in-vitro cross-reactivity with HIT antibodies. Enoxaparin, dalteparin, tinzaparin are ALL contraindicated. |
| Warfarin transition | Start ONLY when platelets ≥ 150K | PO | Overlap argatroban for ≥ 5 days + INR ≥ 2 for 2 consecutive days. Starting warfarin too early depletes protein C Warfarin Limb Gangrene Study, 1997 → paradoxical thrombosis (venous limb gangrene, skin necrosis). Warfarin Limb Gangrene Study, 1997 |
| DOAC transition | Per agent dosing | PO | Rivaroxaban or apixaban increasingly used as alternatives to warfarin for long-term anticoagulation post-HIT. Start when platelets recovered. Duration: 3-6 months minimum (or longer if provoked thrombosis). |
📋 On Rounds
Why can't you bridge to warfarin immediately in HIT?
Warfarin inhibits vitamin K-dependent factors (II, VII, IX, X) but also inhibits protein C and S (natural anticoagulants). Protein C has the shortest half-life (~8h) → it drops first when warfarin is started. In HIT, the patient is already in a prothrombotic state. Starting warfarin creates a transient hypercoagulable window (protein C depleted before factors drop) → venous limb gangrene and skin necrosis.
A patient on heparin has a platelet drop. How do you decide if it's HIT vs other causes?
Use the 4T score: (1) Thrombocytopenia -fall > 50% and nadir ≥ 20K scores highest, (2) Timing -days 5–10 after heparin start (or < 1 day if prior heparin in last 30 days) scores highest, (3) Thrombosis -new confirmed thrombosis scores highest, (4) oTher causes -no other explanation scores highest. 4T score 0–3 = low probability (HIT essentially ruled out, NPV > 99%). 4–5 = intermediate → send PF4 antibody.
Why can you NOT start warfarin until platelets recover above 150K in HIT?
In the acute thrombotic phase of HIT, protein C levels are already depleted (consumed by the ongoing thrombotic process). Warfarin further suppresses protein C (vitamin K-dependent, shorter half-life than other clotting factors) → transient hypercoagulable state → microvascular thrombosis → venous limb gangrene or skin necrosis.
What non-heparin anticoagulants can you use in HIT?
Argatroban (first-line): direct thrombin inhibitor, IV drip, hepatically metabolized (reduce dose in liver failure). Titrate to aPTT 1.5-3× baseline. Prolongs INR → makes warfarin transition tricky (need to check INR after holding argatroban for 4h). Bivalirudin: direct thrombin inhibitor, shorter half-life, preferred in PCI setting or renal failure (cleared by plasma proteases, not kidneys).
❓ What are the components of the 4T score?
T hrombocytopenia: > 50% fall + nadir ≥ 20K = 2pts. T iming: day 5-10 (or < 1 day with prior heparin in last 30 days) = 2pts. T hrombosis: new confirmed thrombosis = 2pts. oT her cause: none apparent = 2pts. Score: 0-3 = low (HIT unlikely), 4-5 = intermediate, 6-8 = high probability → start non-heparin anticoag immediately. [4T Score Validation, 2006]
❓ Why is warfarin dangerous if started too early in HIT?
Warfarin depletes protein C faster than procoagulant factors (protein C half-life = 6h vs Factor II = 60h). In the prothrombotic state of HIT, this early protein C drop → paradoxical thrombosis → venous limb gangrene, skin necrosis (especially at fat-rich sites). Start warfarin ONLY when platelets ≥ 150K and overlap with DTI × ≥ 5 days. Warfarin Limb Gangrene Study, 1997
❓ How do you transition from argatroban to warfarin given that argatroban elevates INR?
Argatroban (a direct thrombin inhibitor) falsely elevates INR because it affects the thrombin-dependent step in the PT assay. Strategies: (1) Start warfarin when platelets ≥ 150K while continuing argatroban. (2) Target combined INR > 4 on overlap. (3) Hold argatroban × 4 hours, then check INR -if ≥ 2, the warfarin effect is therapeutic. (4) Some centers use chromogenic Factor X assay instead (not affected by argatroban).
❓ Can you use DOACs in HIT?
DOACs (rivaroxaban, apixaban) are increasingly used as alternatives to warfarin for the ongoing anticoagulation phase after platelet recovery. Advantages: no INR monitoring, no interaction with argatroban, rapid onset. Limited prospective data but growing retrospective evidence supports safety and efficacy. Start after platelets recovered (≥ 150K). Duration: 4 weeks (HIT without thrombosis) or 3-6 months (HIT with thrombosis).
❓ What is rapid-onset HIT and how does it differ from typical HIT?
Rapid-onset HIT: platelet drop within < 24 hours of heparin re-exposure in a patient with prior heparin exposure within the last 100 days. Pre-formed PF4/heparin antibodies cause immediate platelet activation. Key clue: patient was recently hospitalized or had surgery and now presents with acute thrombocytopenia upon heparin re-exposure. Contrast with typical HIT (day 5-10 onset in heparin-naive patients).
Clinical Examples
📋 Case 1, Post-Surgical HIT with DVT
Patient: 68M, PMH HTN, DM2. POD 8 from R total knee arthroplasty on enoxaparin prophylaxis. Nursing reports new R calf swelling and pain.
Key findings: Platelets dropped from 245K to 88K (64% decline) over 3 days. Duplex US: acute R popliteal DVT. 4T score: 7 (high probability).
Management:
- Stop ALL heparin products immediately (enoxaparin, flushes, line locks)
- Start argatroban 2 mcg/kg/min IV, titrate to aPTT 1.5-3x baseline
- Send PF4/heparin antibody (ELISA) and SRA (confirmatory)
- Do NOT start warfarin until platelets ≥ 150K; overlap with argatroban ≥ 5 days
- Anticoagulate for 3-6 months (HIT with thrombosis = HITT)
Teaching point: LMWH cross-reacts with HIT antibodies in ~90% of cases. A patient with HIT on UFH should NOT be switched to enoxaparin. Always use a non-heparin anticoagulant (argatroban, bivalirudin, fondaparinux).
📋 Case 2, Rapid-Onset HIT
Patient: 55F, PMH breast cancer s/p port placement 3 weeks ago (received heparin flushes). Admitted for chemo, started on UFH DVT prophylaxis. Platelets drop from 190K to 45K within 12 hours.
Key findings: Rapid platelet decline within 24h of heparin re-exposure. Prior heparin exposure < 30 days ago. 4T score: 6 (high).
Management:
- Stop heparin immediately
- Start argatroban; consider bivalirudin if hepatic dysfunction
- Bilateral LE duplex (30-50% have occult DVT)
- Send PF4 Ab + SRA
Teaching point: Rapid-onset HIT occurs within < 24h of re-exposure in patients with pre-formed antibodies from heparin exposure within the last 100 days. The 4T score timing criteria awards 2 points for platelet fall < 1 day with prior heparin exposure.
📋 Case 3, Intermediate 4T Score with Low Clinical Suspicion
Patient: 72M in MICU on UFH for AF. Day 6, platelets drop from 180K to 105K (42% decline). Patient is septic from pneumonia with new pressor requirement.
Key findings: 4T score: 4 (intermediate). Sepsis is a common cause of thrombocytopenia in the ICU. No new thrombosis.
Management:
- Send PF4 antibody (ELISA) given intermediate probability
- If PF4 ELISA negative (OD < 0.40) → HIT essentially excluded (NPV > 99%)
- If PF4 positive → stop heparin, start argatroban, send SRA for confirmation
- Consider alternative causes: sepsis-associated thrombocytopenia, drug-induced (vancomycin, linezolid), dilutional
Teaching point: The 4T score's greatest value is its NPV at low scores (0-3), which essentially rules out HIT. At intermediate scores (4-5), the PF4 ELISA helps decide next steps. Sepsis is the #1 cause of thrombocytopenia in the ICU, not HIT.
📣 Sample Presentation
One-Liner
"Mrs. Liu is a 64-year-old post-op day 7 from hip replacement on heparin prophylaxis whose platelets dropped from 220K to 82K (62% decline). No bleeding. 4T score 6 (high probability). PF4 antibody sent."
Key Points to Cover on Rounds
High-probability HIT (4T score 6: >50% drop, day 5-10 timing, no thrombosis yet, no other cause). Immediate actions: (1) ALL heparin stopped (drips, flushes, line locks), (2) argatroban started at 2 mcg/kg/min, titrate to aPTT 1.5-3× baseline, (3) bilateral LE duplex ordered (30-50% have occult DVT). PF4 Ab pending, SRA pending. No warfarin until plt >150K (protein C depletion risk → skin necrosis, venous limb gangrene). Platelet transfusion NOT indicated. Plan: argatroban until plt recovery, then transition to warfarin with ≥5 days overlap.
Monitoring
- Platelet count daily -should begin recovering within 4-10 days of stopping heparin + starting alternative anticoagulation. If NOT recovering → reconsider diagnosis or check for new thrombosis.
- aPTT q6h while on argatroban -target 1.5-3× baseline. Avoid supratherapeutic levels (bleeding risk).
- INR for warfarin transition -argatroban falsely elevates INR. Check INR after holding argatroban × 4h to get true INR. Some centers use chromogenic factor X assay instead.
- Bilateral LE duplex US at diagnosis -even if no symptoms (30-50% occult DVT)
- Clinical assessment for new thrombosis daily -arterial (stroke, limb ischemia, MI) and venous (DVT, PE). HIT is a prothrombotic state -thrombosis risk highest in first 30 days.
- Skin exam -skin necrosis at injection sites (heparin-induced skin necrosis occurs before overt HIT)
- Duration of anticoagulation: minimum 4 weeks if HIT without thrombosis. 3-6 months if HIT with thrombosis (HIT-T). Some experts recommend extended anticoagulation for arterial HIT events.
⚡ Summary
Summary
The Platelets Drop but the Patient Clots
Anti-PF4/heparin antibodies activate platelets, so HIT is a prothrombotic state despite the thrombocytopenia. Thrombosis, not bleeding, is what harms these patients, and up to half develop it. That single fact governs every management decision on this page.
Recognize the Timing and Magnitude
A platelet fall of more than 50% from baseline, typically 5 to 10 days after heparin exposure, or within hours if there has been exposure in the previous 100 days. The nadir is usually 40,000 to 80,000, rarely below 20,000, so a profoundly low count points to another cause.
Score It With 4Ts Before Testing
Thrombocytopenia magnitude, Timing, Thrombosis or other sequelae, and oTher causes excluded. A low score has a high negative predictive value and effectively rules HIT out, which matters because the alternative anticoagulants are expensive and carry bleeding risk. Intermediate or high scores warrant testing and empiric action.
Stop All Heparin Immediately
IV, subcutaneous, line flushes, heparin-coated catheters and heparin-bonded lines. Even trace exposure perpetuates the immune response. LMWH cross-reacts and is not an alternative. Search the chart actively, because flushes and coated devices are the exposures that get missed.
Stopping Is Not Enough: Start a Non-Heparin Anticoagulant
Argatroban or bivalirudin, or fondaparinux in selected patients. Simply withholding heparin leaves an untreated prothrombotic state, and thrombosis continues to develop. Argatroban is hepatically cleared, so it is the choice in renal failure, and it prolongs the INR, which complicates later warfarin transition.
Do Not Start Warfarin Until Platelets Recover
Wait until the count is at or above 150,000. Early warfarin depletes protein C faster than the procoagulant factors, causing venous limb gangrene and skin necrosis in a patient who is already hypercoagulable. Overlap with the non-heparin agent for at least 5 days and until the INR is therapeutic.
Do Not Transfuse Platelets
Unless there is life-threatening bleeding. Transfused platelets are activated by the HIT antibodies and add fuel to the thrombosis. The low count is not the problem; the clotting is.
Confirm, Document and Duration
Screen with an immunoassay (sensitive, not specific) and confirm with a functional assay such as the serotonin release assay, which is what distinguishes true HIT from the common asymptomatic antibody. Anticoagulate for at least 4 weeks without thrombosis and 3 months with it. Label the allergy prominently in the record, since re-exposure years later can cause rapid-onset HIT and the patient will not know to warn anyone.
📄 One Pager
Text version
Hematology · One Pager
HIT -Heparin-Induced Thrombocytopenia
4T score ≥ 6 → stop ALL heparin + start argatroban. Bilateral LE duplex (30-50% occult DVT). No warfarin until plt ≥ 150K. No LMWH (90% cross-reactivity).
Diagnosis
4T score: Thrombocytopenia > 50% fall, Timing days 5-10, Thrombosis, oTher causes excluded. Score ≥ 6 = high probability → treat immediately. Send PF4 Ab + SRA.
Treatment
STOP all heparin (drips, flushes, line locks). Start argatroban 2 mcg/kg/min IV (reduce in liver failure). Bilateral LE duplex US (30-50% have occult DVT).
Warfarin Rules
Do NOT start until plt ≥ 150K (protein C depletion → venous limb gangrene). Overlap argatroban ≥ 5 days. DOACs emerging as option after acute phase but limited evidence.
Key Drugs
Argatroban2 mcg/kg/min IV drip
BivalirudinAlternative DTI
FondaparinuxSQ (off-label for HIT)
NO LMWH90% cross-reactivity
Pitfalls
- Not stopping ALL heparin (including flushes and line locks)
- Warfarin before plt ≥ 150K (limb gangrene risk)
- LMWH for HIT (cross-reacts in 90%)
- Not checking bilateral LE duplex (30-50% occult DVT)
Related Topics
Acute Leukemia (AML / ALL)AmyloidosisAnemia WorkupAnemia WorkupAnticoagulation ManagementChemotherapy Toxicities