Transitioning from argatroban to warfarin in heparin-induced thrombocytopenia: An analysis of outcomes in patients with elevated international normalized ratio (INR)

Background: Heparin-induced thrombocytopenia (HIT) can lead to catastrophic thromboembolic complications and requires treatment with an alternative, rapidly active anticoagulant, such as a direct thrombin inhibitor (DTI), either to prevent or treat these complications. Switching to oral warfarin after initial treatment with a DTI is necessary in most patients. Most references related to warfarin suggest that an increased risk for bleeding will occur with elevated international normalized ratios (INRs) > 4.6. In patients receiving argatroban, it is not uncommon to achieve an INR > 4 during this transition. Because the clinical outcomes in patients achieving an INR > 4 during combined argatroban/warfarin therapies for HIT are not well described, we evaluated the clinical outcomes of 111 patients with this phenomenon. Methods: We identified patients from the prospective studies of argatroban anticoagulation, Argatroban-911 and Argatroban-915. Data collected from these studies included death from all causes, amputation, new thrombosis, major bleeding, INR values, argatroban doses, aPTT values, platelet counts, and duration of therapy. Results: Patients were on argatroban monotherapy for a median of 2.8 (0.1-8.1) days, and on cotherapy for a median of 3.7 (0.9-12.8) days. The median platelet count was 70.9 (18-325) x 10<Superscript>9supercript stop/L at the time of HIT diagnosis and increased to 94 (30-324) x 10(9)/L by the time warfarin was initiated. At a median argatroban dose of 1.4 (0.2-2.0) mcg/kg/min, the maximum INR ranged from 4.1 to 21.2 (median 6.4, n = 111) and the corresponding aPTT ranged from 48.1 to 105 (median 71, n = 93) seconds. After argatroban cessation, the first recorded INR within 4 to 24 hours ranged from 1.5 to 12.5 (median 2.9, n = 58). Adverse clinical outcomes occurred in 9 (8.1%) patients during cotherapy and in 12 (10.8%) patients after argatroban anticoagulation was discontinued. Adverse clinical outcomes included 7 cases of new thrombosis, 3 amputations, 12 deaths and 1 major bleed. Eleven of 12 (91.7%) patients died due to causes other than thrombosis, and most deaths (83%) occurred following cotherapy. Five (4.5%) patients developed new thrombosis during argatroban/warfarin cotherapy despite an INR > 4. In contrast only 1 (0.9%) patient experienced major bleeding. Conclusion: In patients receiving argatroban/warfarin cotherapy and with an elevated INR > 4, the risk for thrombosis exceeds the risk of bleeding. Traditional paradigms concerning elevated INRs and warfarin may need to be redesigned for the patient population on cotherapy with direct thrombin inhibitors.