Heparin-induced thrombocytopenia (HIT) and thrombosis is characterized by a progressive thrombocytopenia that most often occurs after 5–10 days of heparin exposure. Patients with a history of HIT requiring anticoagulation for cardiopulmonary bypass (CPB) may face devastating consequences unless alternative anticoagulants are used. Management options have included postponing surgery until a time at which the antibody has disappeared and the patient's platelets no longer aggregate after exposure to heparin. Alternative pharmacologic means of anticoagulation include the use of low molecular weight heparin (LMWH), ancrod, warfarin, platelet inhibitors such as aspirin and dextran, as well as the thrombin inhibitors, hirudin or argatroban. [1–4]In addition, other drugs with antiplatelet activity, such as the prostacyclin analogue, iloprost, have been administered before heparin administration in an attempt to prevent platelet aggregation. 
The investigational heparinoid, ORG 10172 (Organon International, The Netherlands), derived from porcine intestinal mucosa, is composed of a heterogeneous mixture of dermatan sulfate, heparan, and chondroitin sulfate. This heparinoid has been an effective anticoagulant for CPB in dogs and has also been used successfully as an antithrombotic agent in patients with HIT. Patients with HIT have decreased antibody cross reactivity with ORG 10172. In humans, case reports have been presented by Doherty et al. and Rowlings et al. in which ORG 10172 was used as the anticoagulant for CPB. In this report, we describe thrombosis that occurred in a patient with HIT receiving ORG 10172 during CPB.
A 72-yr-old man was treated initially with unfractionated porcine heparin and tissue plasminogen activator (t-PA) therapy after an acute anterior wall myocardial infarction. At admission, his platelet count was 301,000/ul (normal, 150,000–400,000/ul). His platelet count at admission to our institution was 75,000/ul. Cardiac catheterization revealed severe three-vessel coronary disease with normal left ventricular function that required coronary artery bypass grafting. Before surgery, while in the coronary care unit, he again experienced severe angina, bradyarrhythmias, hemodynamic instability, and congestive heart failure that required tracheal intubation and placement of an intraaortic balloon pump. Less than 24 h after admission, his platelet count had decreased to 29,000/ul and his heparin was discontinued after HIT was suspected. The patient's plasma was assayed for heparin-associated platelet antibodies to both unfractionated heparin and the LMWH, enoxaparin, using an aggregation method with fresh type O donor platelets and heparin at 0, 0.3, 0.5, and 0.8 U/ml. A positive aggregation test was observed for both agents. Because surgery was required, Organon 10172 was obtained for compassionate use. The patient's plasma demonstrated no aggregation with donor platelets to Organon 10172.
The patient arrived in the operating room tracheally intubated, sedated, and paralyzed, with a intraaortic balloon pump in place via the femoral artery and functional. An arterial catheter and a nonheparin bound pulmonary artery catheter were placed for additional monitoring. The patient received 0.4 mg scopolamine, 10 mg diazepam, and 1 mg fentanyl intravenously for anesthesia, and pancuronium for paralysis. Before incision, baseline coagulation tests revealed: platelet count of 64,000/ul, heparin assay by factor Xa inhibition at < 0.2 units (U)/ml, activated clotting time (ACT) of 132 s (normal 110–140 s), and fibrinogen concentration of 150 mg/dl (normal 200–400 mg/dl).
After sternotomy, before CPB, 5,000 units of ORG 10172 were given intravenously over 2–3 min. An additional 7,500 units of the drug were added to the pump prime before initiation of CPB. During CPB, the patient was cooled to 32 degrees C. Times and laboratory values of platelets, ACT, fibrinogen, and anti-factor Xa activity are listed in Table 1. An additional 15,000 units of ORG 10172 were administered during CPB to maintain an antifactor Xa activity of 1.25 U/ml, which was similar to that used by Doherty et al. 
Forty minutes after initiation of CPB, clots were noted in the surgical field and arterial filter of the CPB circuit. The antifactor Xa activity at this time was 1.75 U/ml. The last proximal coronary graft anastomosis was quickly completed and the patient was promptly separated from CPB. Protamine was not given, because it does not reverse the effects of ORG 10172. 
Thirty minutes after chest closure, diffuse bleeding was noted, and three units of packed red blood cells were given to treat a hematocrit of 23%, and 12 units of platelets were given for a platelet count of 43,000/ul. The patient was subsequently transported to the intensive care unit, but required reexploration for continued bleeding 3 h later for chest tube drainage in excess of 300 ml/h. The antifactor Xa activity at this time was 1.60 U/ml, and the blood temperature was 37.3 degrees Celsius. No surgical bleeding sites were identified intraoperatively. After reexploration, the patient continued to bleed and was given an additional 18 units of platelets and 4 units of packed red blood cells. Bleeding resolved completely 16 h after initial exposure to the anticoagulant. The patient remained hemodynamically stable, allowing his intraaortic balloon pump to be removed and his sedation discontinued on the second postoperative day. On the third postoperative day, he was arousable to noxious stimuli, but he would not follow commands and he continued to require controlled ventilation. A computed tomography scan and electroencephalogram were consistent with a diffuse metabolic encephalopathy. There was no evidence of an intracerebral bleed or infarction. His neurologic status gradually improved, and the trachea was extubated on postoperative day 12, when he was able to follow commands. After discharge to the regular nursing floor, sepsis developed, which led to a protracted recovery. His neurologic status eventually returned to baseline, and he was discharged to an assisted care facility 7 weeks after initial surgery.
Low molecular weight heparin and heparinoids have molecular weights from 4,000 to 6,000 d and may be extracted from bovine or porcine intestinal mucosa by a variety of methods, including gel filtration and ultrafiltration, followed by degradation techniques such as enzymatic or chemical depolymerization. Although LMWH and heparinoids have a lower molecular weight compared with unfractionated heparin, their biochemical structures are similar. They have a greater inhibitory effect on factor Xa than on thrombin and have less platelet interaction when compared with unfractionated heparin. These agents demonstrated promise as antithrombotic agents with minimal hemorrhagic properties.
ORG 10172 is a mixture of sulphated glycosaminoglycans derived from porcine intestinal mucosa. It has minimal antithrombin activity and primarily has anti-Xa activity. Consequently, Organon 10172 only moderately prolongs the activated partial thromboplastin time and ACT. Meulman et al. showed no significant induction of bleeding and a lack of deposition of thrombi with large doses of ORG 10172, suggesting that this drug has no effect on platelet adhesion. This lack of effect on platelets is supported by the finding that ORG 10172 does not affect collagen-induced serotonin release. Ofosu et al. recently demonstrated that ORG 10172 could effectively delay the onset of activation of factors IX and X nearly as effectively as heparin, contributing to its antithrombotic activity.
One of the major complications of heparin administration is thrombocytopenia. Heparin-induced thrombocytopenia is suspected when the platelet count decreases to less than 100,000/ul or the patient experiences new thrombo-hemorrhagic events or increasing heparin requirements while receiving heparin. The mechanism is thought to be immunologically mediated, because platelet-associated immunoglobulin G antibodies that bind to complexes of heparin and platelet factor 4 have been identified in most of these patients. One method of diagnosing HIT is by in vitro platelet aggregation studies, which usually become positive from 6 to 21 days after initial heparin exposure. This test may become positive more rapidly if the patient has had previous exposure to heparin. The etiology of the rapid response is thought to be an anamnestic antibody response after reexposure to heparin. 
The profound anticoagulation required during CPB presents a challenging problem for the clinician when a patient presents with HIT. Low molecular weight heparin and heparinoid compounds are two of the suggested alternatives to standard heparin for anticoagulation during CPB, although they are not currently approved by the Food and Drug Administration for this application, and only a few case reports are published. Although LMWH may affect platelet function less readily than unfractionated heparin, development of progressive thrombocytopenia and positive heparin-associated platelet antibodies have been reported in recent studies. [14,15]In addition, in patients with heparin antibodies to unfractionated heparin, the cross-reactivity with LMWH has been reported to be as high as 80–90%. Crossreactivity of heparin-platelet antibodies with the heparinoid compound ORG 10172 is less (10–18%), such that it may prove useful in patients with HIT. [16–18]Doherty et al. reported the successful use of ORG 10172 as the sole anticoagulant in humans during CPB while maintaining an anti-factor Xa level greater than 1.3 U/ml. Similarly, reports by Mikhailidis et al. and Ortel et al. demonstrated safe use of ORG 10172 in patients with hyperaggregable platelets and a decrease in platelet adhesiveness with documented HIT requiring anticoagulation. Henny et al. also demonstrated the safe use of Organon 10172 in a series of dogs undergoing CPB. Based on the available literature and after a lack of in vitro aggregation of donor platelets with our patient's plasma in the presence of ORG 10172, the drug was used as the anticoagulant for CPB.
Despite maintaining antifactor Xa activity greater than 1.7 U/ml, thrombosis was observed during placement of the proximal grafts toward the end of CPB. This contrasts with the experience of Doherty et al., who reported a satisfactory outcome with antifactor Xa activity of approximately 1.6 units/ml. It is not stated whether Doherty et al. used hypothermic CPB, and the question arises about the anticoagulant activity of ORG 10172 with cooling. There was some correlation between ACT and antifactor Xa activity, but the significance of this is unknown and will probably need further study before establishing a safe ACT for CPB when ORG 10172 is used for anticoagulation. When using unfractionated heparin, full anticoagulation is confirmed by obtaining an ACT > 480 s. This level of anticoagulation is associated with anti-Xa activity of 3.0 U/ml or greater and may warrant careful consideration when using ORG 10172 for CPB.
Currently, there is no reversal agent for Organon 10172, which is a disadvantage that can result in continuous postoperatively bleeding. The administration of protamine has not been associated with decreased anti-Xa activity, and its use cannot currently be recommended. Until a specific reversal agent is available, support of hemostasis by the judicious use of blood products seems warranted, because ORG 10172's half life is approximately 19 h.
In conclusion, even though ORG 10172 has been used successfully in the treatment of HIT, deep venous thrombosis, stroke, and as the sole anticoagulant for patients who require CPB, this case suggests that dosing regimens are not uniform, and that thrombosis is still a potential complication. If ORG 10172 is to be considered a suitable alternative for anticoagulation for patients with HIT, further studies are required to establish proper protocols to facilitate CPB and to establish anti-factor Xa activity above which thrombosis should not occur.