Get PulmCCM’s Weekly Email Update
Stay up-to-date in pulmonary and critical care. No spam.
Heparin-induced thrombocytopenia (HIT) is a dangerous complication of heparin exposure. In HIT, the body creates an autoantibody against the complex of platelet factor 4 bound to heparin. The anti-PF4 autoantibodies can activate platelets and cause life- and limb-threatening thrombosis in arteries and veins.
Platelets generally fall by more than 50% five to ten days after heparin exposure in HIT (or sooner in previously heparin-exposed patients), usually to <150,000/μl.
Thrombocytopenia results in up to 5% of patients exposed to heparin, and most often this is not clinical heparin-induced thrombocytopenia: no antibodies are formed, the platelet drop is mild (about 100,000/μl) and is of no clinical consequence. In these cases, heparin may be continued safely, and platelets often return to normal despite ongoing heparin use.
Some experts have proposed a classification of heparin-induced thrombocytopenia into two types, the life-threatening syndrome being "type II" and the benign being "type I." This scheme may be ill-advised as it encourages conflation and confusion between two profoundly different clinical syndromes, one of which can be lethal. For the remainder of this review, HIT refers to the potentially life-threatening thrombogenic complication of heparin exposure.
Timing and magnitude of thrombocytopenia can be suggestive of HIT, but not diagnostic. Platelet counts usually fall 5-10 days after heparin is begun, but the decline may occur weeks later, even after hospital discharge on outpatient therapy. Nor is there a reliable absolute platelet count for a diagnosis. Most patients have a >50% drop in platelet count, to an average low of 60,000, and rarely below 20,000. However, up to 15% of patients with clinical HIT may remain in the normal range (>150,000) after their decline in platelet count.
Surgical patients appear more susceptible than medical patients. Unfractionated heparin (UFH) appears more likely to cause HIT than low-molecular weight heparin (LMWH) among patients undergoing surgery. However, this increased risk from UFH has not been confirmed in medical patients.
A large proportion (20 to 60%) of patients with HIT experience thrombosis, more often venous than arterial. Deep venous thrombosis, myocardial infarction, stroke, limb gangrene or acute ischemia, and skin necrosis at heparin injection sites are commonly reported. Anaphylaxis is also possible, and may occur with normal platelet counts.
The anti-PF4 antibodies associated with HIT may persist for months or years, and may signify continued vulnerability to thrombosis after heparin administration. Patients with clinical HIT should avoid heparin for life. However, not all patients with positive testing for HIT antibodies have HIT: anti-PF4 antibodies can also be present in the absence of any history or evidence of clinical HIT, and many such patients receive heparin without complication. Other autoantibodies not causing the HIT syndrome (resulting after heparin exposure or from autoimmune disease) can also cross-react with anti-PF4 assays.
Patients with suspected HIT must not receive any additional heparin, and should receive a non-heparin anticoagulant such as argatroban or fondaparinux until HIT can be ruled out. Mortality from thrombosis in HIT has been reported as high as 20%, but is much lower with prompt recognition and treatment.
Diagnosis of HIT
Thrombosis can be catastrophic and preventable, and waiting for lab test results may impose dangerous delays in treatment. A diagnosis of HIT should thus be made presumptively and empiric treatment initiated in patients with moderate to high probability.
Experts advise the 4T score as a HIT screen. The 4T score assesses the timing of platelet decline (5-10 days most often), its magnitude (usually >50%), thrombosis events, and alternate etiologies. A low probability 4Ts score has good negative predictive value in prospective and retrospective validation cohorts; some experts suggest deferring antibody testing (and continuing heparin) for patients with low 4T scores.
In patients with 4T score of moderate or high probability, or where there is strong clinical suspicion despite a low 4T score, all heparin should be discontinued (including any heparin flush) and HIT antibody testing should be performed. While waiting for HIT test results, empiric treatment with a non-heparin anticoagulant is usually advised.
Lab Tests for HIT
Two main test technologies are used to diagnose HIT:
- Immunoassays (ELISA et al) can identify the anti-PF4 HIT antibody in serum. Immunoassays provide fast results and are widely available, but have high false positive and false negative rates. A result of optical density (OD) <0.40 is >99% likely to be a true negative; ODs >2.0 are >90% likely to be true positives; values 0.41-2.00 are intermediate probability (using functional tests as the gold standard).
- Functional tests (serotonin release assays or heparin-induced platelet aggregation) involve mixing patient serum with test platelets. Activation of platelets diagnoses HIT with greater accuracy than immunoassays, but results usually require days.
If a serotonin release assay can be done with a rapid turnaround time, it may be preferred over the lower-accuracy immunoassay.
Immunoassay OD values <0.40 are used by some experts to rule out HIT, and OD > 2.00 to make the diagnosis. For indeterminate values, an empiric non-heparin anticoagulant is often advised while waiting for serotonin release assay results.
Treatment of HIT
For patients with suspected HIT, an anticoagulant is usually advised with a non-heparin agent while awaiting test results:
- Argatroban is an intravenous direct thrombin inhibitor metabolized by the liver; it may be used in patients with renal failure. Argatroban elevates the PTT along with PT/INR. Dosing is targeted to maintain PTT at 1.5 to 3 times baseline. Dose adjustment is required in patients with liver failure. Evidence supports the use of argatroban in HIT.
- Fondaparinux is a synthetic heparin-like analogue given subcutaneously and renally cleared. Monitoring is not required except during long periods of administration, or in patients with renal failure. Little evidence exists for the use of fondaparinux in HIT.
Patients with confirmed HIT should receive an anticoagulant for months after diagnosis; warfarin is the preferred oral agent. Warfarin monotherapy can cause limb ischemia and necrosis, and should be initiated at a low dose (e.g., 5 mg/day) only after effective parenteral anticoagulation is in place and platelet counts have recovered to the normal range. Because argatroban elevates PT/INR, pharmacy consultation is advised in transitioning patients from argatroban to warfarin.
The ideal length of HIT treatment with warfarin is unknown. Antibodies tend to persist for 2-3 months in most people, justifying treatment courses at least this long. Longer treatment courses (e.g., 6 months) have been advised for patients who have had thrombosis events. All patients with a history of confirmed HIT should avoid heparin for life and have a heparin allergy added to their medical record.
Benjamin S. Salter et al. Heparin-Induced Thrombocytopenia: A Comprehensive Clinical Review. Journal of the American College of Cardiology. Volume 67, Issue 21, May 2016.
Greinacher A. Heparin-induced thrombocytopenia. N Engl J Med 2015; 373:252-261. July 16, 2015. DOI: 10.1056/NEJMcp1411910.
Lee GM, Arepally GM. Diagnosis and management of heparin-induced thrombocytopenia. Hematol Oncol Clin North Am. 2013 Jun;27(3):541-63.