HEPARIN-induced thrombocytopenia (HIT) and heparin-induced thrombocytopenia with thrombosis (HITT) are rare but potentially fatal complications of heparin therapy. 1,2Administration of heparin in patients with HITT causes platelet aggregation, thromboembolism, and thrombocytopenia. Therefore, an alternative anticoagulant to heparin is recommended in these patients. Recombinant hirudin (r-hirudin; Refludan, Hoechst Marion Roussel, Inc., Kansas City, MO) is a direct thrombin inhibitor that has been used in patients with HIT and HITT, and case reports have described its use as an anticoagulant in patients requiring cardiac surgery. 3–5The ecarin clotting time (ECT) is probably the best marker for anticoagulation with r-hirudin, 6but this test is not widely available. Furthermore, little information is available on appropriate dosing for patients with renal insufficiency, obesity, or those requiring prolonged cardiopulmonary bypass (CPB). We describe the use of r-hirudin as an anticoagulant in two patients requiring CPB, the first with normal renal function and the second with renal insufficiency. The different dose regimens and the use of activated partial thromboplastin time (aPTT) for monitoring anticoagulation are discussed.
A 59-yr-old, 103-kg man with unstable angina after a non–Q wave myocardial infarction was scheduled for a coronary artery bypass grafting procedure. He had been diagnosed with HITT 6 yr previously at an outside hospital, and although the platelet factor 4 enzyme-linked immunosorbent assay was negative on this admission, his clinical history was significant for thrombocytopenia and thrombosis after heparin exposure. Routine intraoperative monitors were placed, and anesthesia was induced with 5 mg midazolam, 50 μg sufentanil, and 100 mg rocuronium and maintained with intermittent bolus doses of the same drugs. Anticoagulation for CPB was achieved with a bolus dose of 25 mg r-hirudin (0.25 mg/kg) before initiation of CPB, followed by an infusion of 0.15 mg · kg−1· h−1. In addition, 25 mg r-hirudin was added to the extracorporeal circuit prime. During CPB, the aPTT was monitored every 15 min, and it remained >100 s throughout bypass. However, the perfusionist reported an increase in the blood viscosity during rewarming, as noted by visual inspection of the filter screen of the hard shell venous reservoir, although there was no evidence of clot formation. The patient tolerated separation from CPB on the first attempt without inotropic support, with the total CPB time of 79 min. The r-hirudin infusion was discontinued after separation from CPB, and 5 min later the blood in the extracorporeal circuit was completely clotted. The surgery was completed without any complications, and the patient was transferred to the intensive care unit in stable condition. The aPTT values, blood products administered, and postoperative chest tube output are shown in table 1. The patient was discharged from the intensive care unit on the first postoperative day and discharged home on the ninth postoperative day.
Our second patient was a 62-yr-old, 143-kg man with severe mitral valve regurgitation scheduled for a repeat mitral valve replacement. He had undergone coronary artery bypass grafting and mitral valve replacement 71 days previously that was complicated by prolonged postoperative mechanical ventilation, prolonged inotropic support, and renal insufficiency (serum creatinine, 2.2 mg/ml). In addition, he developed deep vein thrombosis that was initially treated with heparin. The heparin was discontinued because of the development of HITT (platelet count of 36,000/ml and thrombosis of the left axillary, subclavian, popliteal, and common femoral veins), which was confirmed by the heparin-induced platelet activation assay. He was then anticoagulated with 22 mg r-hirudin followed by a variable infusion titrated to an aPTT ratio of 2.5 until resolution of the thrombi. Because he continued to have multiple episodes of congestive heart failure and severe mitral regurgitation, he was returned to the operating room.
Routine intraoperative monitors were placed, and anesthesia was induced with 3.5 μg/kg fentanyl, 0.03 mg/kg midazolam, and 100 mg rocuronium and maintained with desflurane, fentanyl infusion, midazolam, and pancuronium. Anticoagulation for CPB was achieved with a bolus dose of 25 mg r-hirudin administered approximately 30 min before initiation of CPB. In addition, 25 mg r-hirudin was added to the extracorporeal circuit prime, and an infusion was started at 0.05 mg · kg−1· h−1with initiation of CPB. The aPTT values were monitored every 15 min. Approximately 135 min after the initial bolus dose, the hirudin infusion was decreased to 0.02 mg · kg−1· h−1, and 90 min later, the infusion was discontinued because as the aPTT values were > 200 s. There were no clinical signs of excessive bleeding (e.g. , bleeding from venous puncture sites) noted during the operation. Separation from CPB was extremely difficult because of a long bypass time (297 min) and poor left ventricular function. Epinephrine 0.5 μg · kg−1· min−1and milrinone 0.75 μg · kg−1· min−1were required for inotropic support. At the time of chest closure, the surgeons reported that the operative field was dry. The aPTT values, blood products administered, and postoperative chest tube output are shown in table 1. In the postoperative period, the patient continued to have profound cardiovascular compromise refractory to inotropic support and developed generalized multisystem organ failure. He died 24 h postoperatively.
The syndromes of HIT and HITT are now well characterized, and the reported incidence of HIT after full-dose heparin therapy approximates 3%. 1Chong 7recommends basing the diagnosis of HIT on four criteria: (1) thrombocytopenia during heparin therapy; (2) absence of other causes of thrombocytopenia; (3) resolution of thrombocytopenia after discontinuation of heparin; and (4) confirmation of a heparin-dependent platelet antibody by in vitro testing. Approximately 20% of patients with HIT develop HITT, with associated morbidity and mortality rates of 80% and 30%, respectively. 1Because diagnostic tests are unable to predict which patients with HIT will develop HITT, any patient who becomes thrombocytopenic during heparin therapy should be considered at risk for thrombosis. 1
The heparin alternatives available in the United States include danaparoid and r-hirudin. The other anticoagulants, argatroban and ancrod, are still under investigation. Danaparoid has a long half-life, requires a complicated dosing regimen, and cross-reacts with heparin antibody in 5–20% of cases; therefore, r-hirudin is the preferred drug. 8Natural hirudin is produced by the leech Hirudo medichalis and is a highly specific direct inhibitor of thrombin; one molecule of hirudin binds to one molecule of thrombin and blocks its thrombogenic activity. 9Anticoagulant activity of hirudin is independent of antithrombin III or any other plasma cofactors and is not neutralized by platelet-secreted proteins or other plasma components. r-Hirudin is derived from yeast cells. It does not interact with platelets or heparin antibodies, does not have fibrinolytic activity, and has a relatively short half-life that makes chemical neutralization unnecessary.
The elimination half-life of r-hirudin is 30–60 min in patients with normal renal function. 10Because the systemic clearance of r-hirudin is proportional to the glomerular filtration rate or creatinine clearance, the dose of r-hirudin must be adjusted based on renal function. In addition, antihirudin antibodies are formed in 40% of patients, which may increase the anticoagulant effects of r-hirudin in renal failure because of decreased renal elimination of the active antibody–hirudin complexes. Furthermore, r-hirudin cannot be neutralized or dialyzed. In obese patients with a body mass index of 25–30, the use of high doses of r-hirudin seems to have a rebound effect that might be caused by redistribution from its storage in fat deposits. 11
The use of r-hirudin for CPB requires careful monitoring of the degree of anticoagulation. Pötzsch et al. 6evaluated three methods of monitoring r-hirudin anticoagulation: activated clotting time (ACT), aPTT, and ECT. Of these three methods, ECT showed a linear correlation with plasma hirudin concentrations, whereas both ACT and aPTT had poor correlation with plasma hirudin concentrations. 6ECT is based on the fact that snake venom enzyme, ecarin, converts prothrombin to meizothrombin, which has weak coagulant activity and is neutralized by r-hirudin, resulting in a dose-dependent prolongation of clotting times. 12Although ECT is preferable to both ACT and aPTT, it is not widely available, and we were unable to obtain either ECT or plasma hirudin concentrations at our institution. Because both of our patients required urgent surgical management, we had to rely on a less reliable marker of hirudin anticoagulation, aPTT.
The dose of r-hirudin for anticoagulation for thrombosis is 0.4 mg/kg bolus followed by an infusion of 0.15 mg · kg−1· h−1(package insert, Refludan). The dose should not be increased beyond that required for 110 kg of body weight. With CPB, a higher dose is needed to ensure adequate anticoagulation with exposure to the extracorporeal circuit. Pötzsch et al. 6reported successful anticoagulation without thrombotic or bleeding complications using an initial bolus dose of 0.25 mg/kg, 0.2 mg/kg added to the extracorporeal circuit prime, and intermittent bolus doses of 5 mg to maintain plasma hirudin concentrations > 2.5 μg/ml. It is suggested that the infusion rate be adjusted to maintain the aPTT ratio (i.e. , aPTT at the time over an aPTT reference value) of ≥ 2.5. 13Reiss et al. 3reported the successful use of r-hirudin in an aortic valve replacement, in which they monitored anticoagulation with both ECT and aPTT. In this patient, the plasma hirudin concentration was kept between 2.5 and 3.2 μg/ml, ECT was between 300 and 456 s, and aPTT was between 90 and 130 s. With our first patient, although the aPTT ratio was >4 and the aPTT values remained >100 s throughout CPB, the blood viscosity likely increased during rewarming, and the blood clotted in the extracorporeal circuit within 5 min after termination of CPB (possibly suggesting inadequate anticoagulation). There are several explanations for possible inadequate anticoagulation in this patient. Plasma levels of r-hirudin do show a high degree of interindividual variability during continuous infusions, even when the dose is adjusted for body weight. 6Furthermore, the aPTT is a relatively unreliable marker of r-hirudin anticoagulation.
The concerns in our second patient with regard to r-hirudin dosing included renal insufficiency (serum creatinine, 2.2 mg/ml), obesity (body weight, 143 kg), the possibility of prolonged CPB time, and our previous experience with possible inadequate anticoagulation when maintaining aPTT > 100 s. Although the laboratory routinely does not extend an aPTT beyond 100 s, it can be measured to 200 s if requested. The bolus dose of r-hirudin (i.e. , 25 mg) was based on 0.2 mg/kg (up to 110 kg), and the initial infusion rate of 50 μg · kg−1· h−1was based on the serum creatinine levels (as recommended by the manufacturer). The infusion of r-hirudin had to be reduced by more than half of the recommended dosage (20 μg · kg−1· h−1) after approximately 2 h, suggesting cumulation. Although the aPTT values were maintained higher than those suggested, there were no signs of overdose (i.e. , excessive bleeding) in this patient. In addition, we were able to achieve adequate hemostasis after CPB. Therefore, we suggest that if ECT is unavailable and aPTT is being used, the aPTT should be maintained at levels higher than previously recommended in patients undergoing cardiac surgery.
In summary, we observed possible signs of inadequate anticoagulation when using an aPTT ratio > 4 (aPTT > 100 s) to guide infusion of r-hirudin, suggesting that higher levels may be necessary to prevent thrombotic complications during CPB. We also report the use of r-hirudin in a patient with renal insufficiency, obesity, and prolonged CPB time in whom higher aPTT values (approximately 200) provided adequate anticoagulation during CPB without hemorrhagic complications or difficulty in reversal of anticoagulation.