PCI Options in Heparin-Induced Thrombocytopenia

Heparin-induced thrombocytopenia (HIT) is a rather rare complication of heparin therapy, but may occasionally lead to devastating or even life-threatening complications. HIT is more frequent after therapy with unfractionated than low molecular weight heparin. The inherent problem of prothrombosis in HIT can be a major obstacle in patients who need therapy with percutaneous coronary interventions (PCI), especially with stent implantation. This is due to the iatrogenic plaque disruption and promotion of an endovascular thrombosis process during PCI, and the added risk of stent thrombosis. Therefore, it is evident that if the interventional cardiologist does not pay appropriate attention to this syndrome, a major, life-threatening complication may occur in the catheterization laboratory or shortly after the PCI procedure. The low platelet count in HIT is rather deceiving, as the platelets may actually be remarkably reactive to thrombogenic stimuli. Another “difficult” feature of HIT is the inadequacy of laboratory confirmation in a clinically relevant (timely) fashion; thus, HIT is practically viewed as a “clinical” diagnosis, and therapy should be mainly established on clinical grounds. The only clear-cut scenario would be patients with prior HIT, who should be automatically considered at subsequent risk for HIT for life. What options does current pharmacotherapy offer for HIT patients in need for PCI-stent? At best, this is not adequately studied, primarily due to the fact that HIT, especially in combination with thrombosis syndrome, is a rare clinical event. In general, thrombotic events with HIT are treated with a direct thrombin inhibitor infusion until the platelet count recovers or the thrombotic complications resolve. Such agents should be considered for PCI in HIT patients, and three recent reports (two of which presented in this issue of the Journal) have assessed this strategy.1–3 Acute platelet inhibition with a glycoprotein (GP) IIb/IIIa inhibitor is another theoretical option during PCI in HIT patients. Use of these agents in thrombocytopenia appears counter intuitive, but it is based on the fact that platelets are remarkably prothrombotic in HIT, as stated above. It is interesting that these agents were used selectively in combination with all 3 direct thrombin inhibitors.1–3 I would presume this was done for PCI in HIT with thrombosis, or in HIT without thrombosis but with high-risk PCI lesions or clinical scenarios (e.g. acute coronary syndromes). No specific guidelines for the algorithm the GP IIb/IIIa inhibitors were selected is described in any of these studies (“operator choice” is quoted). Due to the absence of any control group and their small sample size, these studies are methodologically limited. One should also note the very slow recruitment rate (especially in the ATBAT study), which is indicative of how rare HIT-PCI cases really are. It is reassuring, though, that cardiac death occurred in only 1 patient (ATBAT study), and under circumstances that do not clearly signify stent thrombosis; the mortality occurred 2 days after LAD stent in the context of bradycardia without ECG changes or chest pain. I would think that an LAD stent thrombosis should have been clinically overt. Although further details are not given, it is possible that a pulmonary embolus or other (noncoronary) HIT-related thrombotic event might have been the cause of death. There was also 1 bleeding death (retroperitoneal hemorrhage) and 1 noncardiac death (sepsis) both in the lepirudin study, in patients also treated with tirofiban and abciximab, respectively. Dosage issues. Despite the study limitations, I should commend the investigators and the study sponsors for attempting to address this complex (and rare) clinical setting. The take-home message from these studies is related intimately to our ability to formulate a crisp, operator-friendly plan for HIT-PCI cases. Although the therapeutic platform of a direct thrombin inhibitor cannot be challenged, the dosage issues remain obscure. Perhaps due to the greater familiarity of the internationalists with large, contemporary trials with bivalirudin in PCI without HIT, the dosage appears more clearly stated in the ATBAT study. It is interesting, however, that the only mortality occurred in the lower of the two dosages studied. Regardless of the reservations stated above regarding the cause of the mortality event, one should raise the question of how long should bivalirudin be used in HIT-PCI cases. Perhaps somewhat longer than PCI in no-HIT cases (which is typically interrupted right after the end of the PCI) should be considered if there is clinical suspicion that the HIT syndrome still requires direct thrombin inhibitor infusion. In other words, one should be mindful that the agent should also be treating another condition in addition to PCI itself. Finally, the bleeding events were rather scarce, especially with the low-dose protocol (0.75 mg/kg bolus plus 1.75 mg/kg/hour infusion for up to 4 hours after PCI). The study with lepirudin used a variety of doses, but I believe only 0.6 mg/kg appears attractive. It is also important that the authors recommend NOT performing PCI unless the platelet count has recovered to at least 50,000/mcL. Additional bolus doses are recommended according to the ACT value. Bleeding events appear more frequent in this study, but it hard to extrapolate between-study differences, given the different patient selection, institutions, and operators. Perhaps the more liberal use of GP IIb/IIIa inhibitors in this study, compared to ATBAT, contributed to bleeding events. In brief, I think these two studies complement earlier reports with argatroban for HIT-PCI cases3,4 and provide additional validated options for PCI in such high-risk cases. It is interesting that the pharmacological properties of these three agents are complementary: argatroban, clearly hepatically; lepirudin, renally; and bivalirudin, both by renal and by catalysis in plasma. Nonetheless, I would hope that these registries should continue to accumulate cases even after approval of these agents, in order to better clarify the treatment of these complex patients in the cardiac catheterization laboratory. I wonder if the strategy of avoiding unfractionated heparin upfront would decrease HIT development in the first place and provide a more clear-cut approach to the clinical difficulties that HIT-PCI poses.

1. Mahaffey KW, Lewis BE, Wildermann NM, et al. The Anticoagulant Therapy with Bivalirudin to Assist in the performance of PCI in patients with heparin-Induced Thrombocytopenia (ATBAT) study. J Invas Cardiol 2003;15:611–616. 2. Cochran K, DeMartini TJ, Lewis BE,et al. Use of lepirudin during percutaneous vascular interventions in patients with heparin-induced thrombocytopenia. J Invas Cardiol 2003;15:617–621. 3. Lewis BE, Matthai WH, Coehn M, et al., for the ARG-216/310/311 Study Investigators. Argatroban anticoagulation during percutaneous coronary interventions in patients with heparin-induced thrombocytopenia. Cathet Cardiovasc Intervent 2002;57:177–184. 4. Matthai WH. Use of argatroban during percutaneous coronary interventions in patients with heparin-induced thrombocytopenia. Semin Thromb Hemost 1999;25:57–60.