Abstract
TO THE EDITOR: Kubiak et al. 1 present a case of a 77-year-old patient with normal hepatic enzymes and metastatic cancer with recurrent deep vein thrombosis (DVT) and a history of heparin-induced thrombocytopenia (HIT) 4 years earlier who was treated with argatroban. Of note were elevated serum sodium 151 mEq/L, blood urea nitrogen 57 mg/dL, and serum creatinine 2.1 mg/dL. Results of other liver function tests, including total bilirubin, were within normal limits. Interestingly, the patient's baseline international normalized ratio (INR) was elevated to 1.9 without any concurrent anticoagulation.
Due to development of a new upper-extremity DVT, an argatroban infusion was initially started at 2 μg/kg/min 1.5 hours after the first dose of warfarin 5 mg was administered. A second warfarin dose was administered on hospital day 2 (19 h later). Elevated activated partial thromboplastin times (aPTTs) and INRs were obtained on hospital days 2, 3, and 4. The argatroban dose was held and eventually titrated downward to 0.25 μg/kg/min before being discontinued on hospital day 3. The INR continued to increase over that time despite a falling aPTT.
Surprisingly, despite the elevated aPTT, fondaparinux 5 mg was administered on hospital days 3 and 4. On hospital day 4, the patient developed hemoptysis; the measured aPTT was 75.1 seconds and INR was 9.6. He received 4 units of fresh frozen plasma (FFP) and 2.4 mg of recombinant activated factor VIIa (rVIIa), resulting in an aPTT of 56.5 seconds and INR of 2.3 that trended upward the next day prior to decreasing. The authors concluded that argatroban was the most likely agent responsible for the continued aPTT and INR elevations. We present several observations and comments regarding the coagulation tests and selection of anticoagulation with fondaparinux.
False positive in vitro laboratory observation of the INR from argatroban has been documented.2–4 Studies to date indicate that a linear correlation exists between argatroban dose and INR that may vary depending on the assay used.2–4 These data suggest that warfarin might have played a more significant role in the observed aPTT and INR values of the case presented here based on increasing INR with minimal changes in aPTT, yielding the first crossover in the INR and aPTT lines. In addition, clotting factors in FFP or use of rVIIa may have a shorter duration of effect than warfarin, leading to an INR rebound and the observed second crossover in aPTT/INR values. 5 The aPTT and INR ratios then followed a similar pattern. Another factor to consider is that warfarin can also yield an increase in aPTT. 6 Noting the fairly rapid decline in aPTT and INR values early in the patient's course upon withholding argatroban, in addition to the >150-second aPTT value being a hemodiluted sample, it is more likely that the prolonged aPTT experienced by this patient was an effect of warfarin administration and was not attributable to argatroban.
Of additional note is that elevated admission laboratory values might also be consistent with dehydration instead of reduced renal function. During this hospital admission, liver metastases were subsequently discovered that may have played a role in the coagulopathy, as mentioned by the authors. The INR that was measured approximately 5 days following argatroban and warfarin discontinuation was 1.9, suggesting a baseline coagulopathy potentially secondary to liver metastases.
This case highlights the need for careful attention to the interpretation of laboratory tests of coagulation in patients receiving direct thrombin inhibitors. An elevated INR at baseline should prompt a complete workup for coagulopathy. While the authors suggest that this elevation may have been due to vitamin K deficiency secondary to antimicrobial use, the safety or protocol of initiating anticoagulants in patients with coagulopathy has not been established. In patients with prolonged elevation of aPTT, initiation of new anticoagulants, such as fondaparinux, should be delayed until the therapeutic effects of other anticoagulants are no longer present. In this case, fondaparinux was initiated in the presence of an elevated aPTT and INR, with resultant hemoptysis. This case report further underscores the importance of understanding the relationship between anticoagulants and the less-specific laboratory indices used to measure their effects.