Heparin-induced thrombocytopenia (HIT) is one of the most serious adverse drug reactions encountered in hospitals (1). Although this immune-mediated condition causes a decrease in platelets, patients suffering from HIT are also at increased risk for thromboses, which can occur in both the arterial and venous systems (2,3).
Obviously, perioperative management of patients having surgical procedures requiring immediate anticoagulation is a challenge in patients with a history of HIT. In the following case, we used argatroban (GlaxoSmithKline Pharmaceuticals, Philadelphia, PA), one of three available direct inhibitors of thrombin, for this purpose. Argatroban’s pharmacokinetic and pharmacodynamic profiles make it ideal for use in a carotid endarterectomy (CEA). The recommended dosage for treatment of HIT consists of commencing the infusion at 2 μg · kg−1 · min−1 followed by increasing the infusion up to 40 μg · kg−1 · min−1 to achieve a target partial thromboplastin time (pTT) of 1.5 to 3 times baseline. An initial loading dose is not required. With this dosing regimen, the steady-state levels of anticoagulation and of the drug concentration are achieved within 1–3 h. Because CEA requires a faster onset of anticoagulation, we administered a bolus of argatroban 150 μg/kg and then started the infusion at 5 μg · kg−1 · min−1. We derived our modified dose using data published by Swan and Hursting (4) evaluating pharmacokinetic and pharmacodynamic effects of argatroban. Argatroban also has a short elimination half-life of 39–51 min (5). The established method of measuring anticoagulant activity is pTT. Specific therapeutic ranges for other clot-based tests are not as well studied, but argatroban will increase prothrombin time (PT), international normalized ratio (INR), and activated clotting time (ACT) in a dose-dependent fashion. Contraindications to using argatroban are recent lumbar puncture, history of a bleeding diathesis or coagulation disorder other than HIT/heparin-induced thrombocytopenia-thrombosis syndrome, or the occurrence of aneurysm or stroke within the last 6 mo (4,6). This is the first case report of the intraoperative use of argatroban for CEA.
A 76-yr-old woman with a history of HIT was scheduled for a left CEA for asymptomatic critical stenosis, as determined by standard duplex criteria. Her medical history was significant for hypertension, chronic renal insufficiency with a creatinine clearance of 20 mL/min, and an abdominal aortic aneurysm repair 3 yr before. After the repair of the aneurysm, she had developed HIT, confirmed by serologic evaluation using 14C serotonin release assay.
The patient was premedicated with 2 mg of midazolam and taken to the operating room where standard monitors and an arterial catheter (normal saline flush) were placed. After anesthetic induction, PT, pTT, INR, and ACT were drawn. Table 1 gives temporal detail of the coagulation profile and the argatroban infusion throughout the case. Argatroban was then administered as a bolus of 8.5 mg (150 μg/kg) followed by an infusion at 5 μg · kg−1 · min−1. Coagulation variables were repeated at 15-min intervals, and the infusion rate was titrated to maintain an ACT at or more than approximately 200 s per cardiovascular protocol at our institution. Adequate anticoagulation was achieved in 29 min at the above dose to allow for cross-clamping of the internal carotid artery (ICA). Because the ICA was extremely redundant, a CEA was performed using an eversion technique with resection and reimplantation of the ICA into the common carotid artery. After completion of the CEA, the infusion was stopped. A residual kink in the ICA above the level of the anastomosis was noted, and an intraoperative duplex ultrasound was performed. The coagulation variables, as seen in Table 1, then spontaneously shifted towards baseline during this period. However, the duplex ultrasound confirmed a high flow velocity at the level of the kink in the ICA. The infusion was recommenced at a rate of 5 μg · kg−1 · min−1. The carotid artery was again cross-clamped and a 2-cm collagen impregnated Dacron patch was applied. Although a minimal amount of platelet debris was removed from the ICA at the level of the kinked ICA, the kink itself was considered to be contributory to both the hemodynamic changes and the platelet aggregate. No intraluminal thrombus was otherwise identified. After completion of the patch angioplasty, adequate surgical hemostasis was achieved within 30 min of the second discontinuation of the infusion. The coagulation variables spontaneously resolved after discontinuation of the infusion. The patient emerged from anesthesia, and the trachea was extubated without complication at the end of the surgery. She had no evidence of thrombosis, electroencephalogram changes, bleeding, or hematoma after surgery. She was discharged on postoperative Day 1 in good condition.
Three direct thrombin inhibitors are available in the United States. Argatroban was Food and Drug Administration approved in June 2000, and its labeled uses are for thrombosis prophylaxis or treatment in patients with HIT and for use in percutaneous coronary intervention in patients with or at risk for HIT. Lepirudin was approved in March 1998, and it is labeled for use in HIT and associated thromboembolic disease to prevent further thromboembolic complications. Finally, bivalirudin was approved December 2000 and is labeled for use with aspirin in patients with unstable angina undergoing angioplasty (7). Bivalirudin definitely deserves mentioning in conjunction with this case, especially because it is beginning to be studied in many situations as an alternative to heparin. For example, a recent trial of 100 subjects undergoing off-pump coronary artery bypass surgery showed no clinically significant difference in blood loss between a heparin and protamine group and a bivalirudin group, and the patients receiving bivalirudin actually showed better graft flow (8). Had we not chosen argatroban, bivalirudin would have been an acceptable option.
There are three main reasons why we believe argatroban was optimal for use in the case presented. First, unlike heparin, argatroban will not precipitate HIT. The direct thrombin inhibitors reversibly bind to the thrombin active site, bypassing interaction with antithrombin III. They do not cross-react with the heparin-induced antibody (3).
Second, argatroban is the only direct thrombin inhibitor that is hepatically eliminated. Furthermore, no adjustment in dosing of argatroban is required for renally impaired patients (4). This presents a major advantage because many patients with a history of HIT also have renal compromise (1). Lepirudin and bivalirudin, both renally eliminated, could potentially accumulate in patients with renal disease, and because there is no antidote for the direct thrombin inhibitors, their use in such patients could lead to unpredictable periods of coagulopathy and increased risk of adverse effects from bleeding. Although bivalirudin is also cleared by proteolysis, its dose still needs to be adjusted in the context of renal failure. Bivalirudin would be a better choice with renal failure than lepirudin because the prolongation of the context-sensitive half-life is not as long (3). Both argatroban and bivalirudin have similar efficacy and safety profiles and short half-lives. However, we consider argatroban a better choice in this case because its use eliminates the unnecessary variable of dose adjustment for renal insufficiency. Of further note, if undesired excessive anticoagulation occurs while using argatroban, discontinuing its administration will result in a return to baseline coagulation variables within two to four hours. In the meantime, the recommended treatment is primarily supportive administration of crystalloids, colloids, or blood products (6).
Third, argatroban’s pharmacokinetic profile, specifically a quick onset and a short half-life, make it useful during CEA. Once the bolus was given and the infusion started, adequate anticoagulation was achieved in time for cross-clamping. The infusion was discontinued after 56 minutes, and only 12 minutes after discontinuation (when duplex revealed a hemodynamically significant problem requiring repair) was it noted that the coagulation variables were recovering. A second rapid onset occurred when the infusion was restarted and was adequate for cross-clamping again, as well as placement of the Dacron patch. The small amount of localized platelet aggregate at the time of re-exploration was attributed to the residual kink in the ICA rather than inadequate anticoagulation. Although a final set of coagulation variables was not obtained after surgery to document complete reversal of anticoagulation, we did obtain values at 15- and 30-minute intervals after cessation of the infusion, which demonstrated a trend toward normalization. Furthermore, it was clinically evident that the coagulopathy was adequately, spontaneously reversed by the end of surgery.
Two practical issues concerning this case are coagulation monitoring and the availability of argatroban. Monitoring of coagulation variables in the operating room continues to be imperfect. Whereas ACT provides quick feedback of the status of coagulation, it is less sensitive and reproducible than PT and pTT, both of which can take up to an hour to process, depending on the institution. Swan and Hursting (4) were able to demonstrate a consistent linear relationship between ACT and pTT during the use of argatroban, and therefore, we believe that ACT is the best coagulation monitor available for monitoring argatroban in the operating room. Argatroban is a formulary drug at our hospital, but obtaining it at some institutions could be challenging. We suggest that it should be available as a nonformulary alternative to lepirudin or bivalirudin if the physician emphasizes the need for a direct thrombin inhibitor not requiring renal adjustment in the setting of HIT with renal insufficiency.
1. Rice L, Nguyen PH, Vann AR. Preventing complications in heparin-induced thrombocytopenia: alternative anticoagulants are improving patient outcomes. Postgraduate Med 2002;112:85–9.
2. Warkentin TE. Heparin-induced thrombocytopenia and the anesthesiologist. Can J Anaesth 2002;49:S36–49.
3. Dager WE, White RH. Treatment of heparin-induced thrombocytopenia. Ann Pharmacotherapy 2002;36:489–503.
4. Swan S, Hursting M. The pharmacokinetics and pharmacodynamics of argatroban: effects of age, gender, and hepatic or renal dysfunction. Pharmacotherapy 2000;20:318–29.
5. McKeage K, Plosker G. Argatroban. Drugs 2001;61:515–22.
6. SmithKline Beecham Pharmaceuticals. Argatroban (package insert). August 2000.
7. Killion KH, Kastrup EK. Drug facts and comparisons. St. Louis: Wolters Kluwer Health, Inc, Updated monthly. July 2004:173–176E.
8. Merry AF, Raudkivi PJ, Middleton NG, et al. Bivalirudin versus heparin and protamine in off-pump coronary artery bypass surgery. Ann Thoracic Surgery 2004;77:925–31.