Approximately 50% of patients with HIT do not have a new HIT-associated thrombosis at the time HIT is first clinically suspected on the basis of thrombocy-topenia alone (Warkentin and Kelton, 1996; Greinacher et al., 1999a,b, 2005). In a retrospective cohort study of 62 such patients with "isolated HIT," the subsequent 30-day cumulative thrombotic event rate was high (52.8%) (see Fig. 2 in Chapter 3). The rate of thrombosis was similar in the two largest patient subgroups: patients treated with discontinuation of heparin therapy alone (20/36, 56%) and patients treated with substitution of warfarin for heparin (10/21, 48%). The majority of events involved the venous circulation (4 : 1 ratio), with six of the 62 patients developing pulmonary embolism (two fatal); another patient who died suddenly may also have had a fatal pulmonary embolism.
In a subsequent large retrospective cohort study of serologically confirmed HIT performed by Wallis and coworkers (1999), a 38% thrombotic event rate was observed in patients with isolated HIT managed by cessation of heparin. Further, early cessation of heparin was not associated with a reduction in the rate of thrombosis. In another study, Zwicker and colleagues (2004) observed that five (36%) of 14 patients with clinically suspected HIT who tested strongly positive (>1.00 units of optical density) for anti-PF4/heparin antibodies by anti-PF4/ heparin enzyme-immunoassay (EIA) developed symptomatic thrombosis. The high symptomatic thrombotic event rates observed in these three retrospective cohort studies are consistent with prospective treatment cohort studies that also observed a high rate of thrombosis (5-10%/day over the first 1-2 days) soon after the diagnosis of HIT (Greinacher et al., 1999a,b, 2000).
These high thrombotic event rates among patients with isolated HIT suggest that many patients may have had subclinical deep-vein thrombosis (DVT) at the time that HIT was first suspected. Indeed, Tardy and colleagues (1999) found that eight of 16 patients identified as having isolated HIT had subclinical DVT identified by systematic duplex ultrasound investigations.
Recommendation. Patients suspected to have acute HIT should undergo imaging studies for lower limb DVT, especially those at highest risk for venous thromboembo-
lism, such as postoperative patients (grade 1C).
There is evidence that therapeutic-dose anticoagulant therapy of isolated HIT is effective. In a retrospective analysis of patients with isolated HIT comparing treatment with danaparoid and lepirudin, it was observed that patients who received prophylactic-dose danaparoid (750 U sc b.i.d. or t.i.d.) had a trend to a higher rate of thrombosis than patients treated with lepirudin (0.1mg/kg b.w./h, aPTT-adjusted) (Farner et al., 2001). In contrast, patients with HIT-associated thrombosis had similar outcomes when treated with therapeutic doses of either drug. This indicates that therapeutic, rather than prophylactic, doses of danaparoid may be more effective for patients with isolated HIT (Farner et al., 2001; Warkentin, 2001). Further evidence supporting the use of therapeutic-dose anticoagulation for isolated HIT includes the results of (aPTT-adjusted) therapy with the DTIs, lepirudin, and argatroban (Lubenow et al., 2004; Lewis et al., 2001, 2003, 2006).
We usually prescribe an alternative anticoagulant in therapeutic doses in this situation of strongly suspected (or confirmed) HIT. However, prophylactic-dose anticoagulation is a reasonable option in patients with a low or intermediate likelihood of having HIT (see Table 7 in Chapter 2) or in HIT patients judged to be at high risk for bleeding complications. Another option could be regular screening for venous thrombosis without anticoagulation in a patient at very high bleeding risk. Thrombocytopenia itself should not be considered a contraindication to anticoagulation in patients with HIT, as petechiae and other spontaneous hemor-rhagic manifestations are not usually seen in these patients (see Chapter 2). However, if the platelet count is less than 20 X 109/L and bleeding signs, but not thrombosis, are observed, then alternative diagnoses such as posttransfusion purpura or other drug-dependent immune thrombocytopenic disorders should be considered (Kiefel, 2004).
Recommendation. Alternative therapeutic-dose anticoagulation with an appropriate anticoagulant, such as danaparoid, lepirudin, or argatroban, should be considered in patients strongly suspected (or confirmed) to have HIT even in the absence of symptomatic thrombosis. Anticoagulation should be continued at least until recovery of the platelet counts to a stable plateau (grade 1C).
It is uncertain whether anticoagulation of isolated HIT beyond the time to platelet count recovery (to a stable plateau) is required, if there are no ongoing risk factors for thrombosis, such as atrial fibrillation or prolonged immobility. In the prospective lepirudin studies (Lubenow et al., 2004), the risk of subsequent thrombosis (35 day follow-up) among patients with isolated HIT treated until full platelet count recovery was low. We therefore do not usually give prolonged anticoagulation in our own clinical practice. (See also sections III.E and VI.B regarding contraindication to the use of vitamin K antagonists during the acute thrombocytopenic phase of HIT). It is reasonable to repeat a duplex ultrasound of the lower extremities prior to discharging a patient to home when ongoing anticoagulation will not be given.
Was this article helpful?