Bivalirudin as a Foundation Anticoagulant in Peripheral Vascular Disease: A Safe and Feasible Alternative for Renal and Iliac In

David E. Allie, MD, Mitchell D. Lirtzman, MD, Charles H. Wyatt, MD, V. Antoine Keller, MD, Mohamed H. Khan, MD, Muhammad A. Khan, MD, Peter S. Fail, MD, Chris J. Hebert, RT, RCIS, Sonja D. Ellis, MBA, Elena Mitran, MD, Gary Chaisson, RT, RCIS, Samuel Stagg, Jr., MD, Adam A. Allie, MS, Craig M. Walker, MD
David E. Allie, MD, Mitchell D. Lirtzman, MD, Charles H. Wyatt, MD, V. Antoine Keller, MD, Mohamed H. Khan, MD, Muhammad A. Khan, MD, Peter S. Fail, MD, Chris J. Hebert, RT, RCIS, Sonja D. Ellis, MBA, Elena Mitran, MD, Gary Chaisson, RT, RCIS, Samuel Stagg, Jr., MD, Adam A. Allie, MS, Craig M. Walker, MD
Over 200,000 peripheral interventions were performed in 1997; this number continues to grow at a rapid pace.1 Peripheral artery disease (PAD), also referred to as peripheral vascular disease (PVD), affects 12–20% of Americans >= 65 years old, resulting in 4.5–7.6 million people with PAD.2 Many of these people will be candidates for percutaneous peripheral intervention (PPI). Unfractionated heparin (UFH) has been the antithrombotic foundation for percutaneous coronary intervention (PCI) and PPI since the first reported angioplasty by Andreas Gruentzig in 1979.3 Despite significant therapeutic and technological advances in both PCI and PPI, ischemic and hemorrhagic complications remain the most commonly associated risks during PCI and are a major source of morbidity, mortality and costs. Major and minor hemorrhagic complications in PCI vary widely depending on the reporting standards and definitions, but still range from 0.4–17%.4 Ischemic complications, measured by the combined clinical endpoint of death, myocardial infarction (MI) or revascularization, continue to be reported at approximately 7.5% in the low-risk PCI patient and can double to over 14.0% in the high-risk, post-MI PCI patient with heparin as anticoagulant.5–7 Event rates in PPI have not been well characterized in the literature. Major complication rates in selected studies published after 1993 range from 0–7.6% for aortoiliac stenosis and occlusion, with 1 year patency rates reported from 68–97%.8 Event rates for renal PPI are even less well defined in the literature. Pharmacological and clinical limitations of UFH, including unpredictable anticoagulant effects, inability to inhibit clot-bound thrombin, and increased risk of heparin-induced thrombocytopenia (HIT), have led to the investigation of a number of direct thrombin inhibitors as alternative anticoagulants.9 Among available direct thrombin inhibitors, bivalirudin (Angiomax®; The Medicines Company, Parsippany, New Jersey) has demonstrated the most promise as an alternative anticoagulant to UFH in PCI. Based on the mechanism of action, pharmacokinetic profile and favorable clinical outcomes in PCI, bivalirudin also holds promise in PPI. This study evaluated the safety and feasibility of bivalirudin in the treatment of renal and iliac PPI compared to an historical UFH control group matched for time, patient demographics, clinical characteristics, technique and operator. According to the documented literature, this is the first report describing the use of bivalirudin and describes a safety and feasibility study designed to clinically evaluate this agent in renal and iliac PPI. Methods Patients presenting with severe renal artery stenosis or with severe iliac artery stenosis and undergoing PPI received as the sole anticoagulant bivalirudin in a 0.75 mg/kg bolus with 1.75 mg/kg continuous infusion for procedural duration. Exclusion criteria included glycoprotein (GP) IIb/IIIa receptor inhibitor use, brachial artery access and serum creatinine >= 1.5 mg/dl. Since this was a feasibility study in PPI, baseline data with bivalirudin were sought in a patient population with relatively normal renal function to avoid the confounding influence of renal impairment on patient outcomes. A retrospective chart review was undertaken by trained research nurses and investigators to obtain the historical UFH control group for renal and iliac PPIs occurring during a similarly matched time frame. Cases were reviewed and matched for patient demographics, clinical characteristics, technique and operator. Heparin was administered at a bolus dose of 50–100 units/kg with target activated clotting time (ACT) > 300 seconds. All patients received preprocedural aspirin 300 mg and clopidogrel 75 mg once daily. Simple descriptive statistics were employed and Chi-square tests were performed on the frequencies in both renal and iliac PPI groups to compare bivalirudin against UFH. Endpoints. The efficacy was evaluated as procedural success, defined as successful balloon angioplasty and stent deployment with reinstitution of flow and = 3 g/dl, cerebrovascular accident, any complications requiring surgery, intracranial bleeding, retroperitoneal hematoma, > 5 cm groin hematoma, or > 2 units packed cell transfusion. Minor vascular access complications included all other non-intracranial or retroperitoneal bleeding, small ( 5 cm non-surgical hematomas. Of the 6 major vascular complications in the UFH group, four were > 5 cm hematomas and 2 required > 2 units packed cell transfusions. Of the 5 minor vascular access complications in the bivalirudin group, three were 5 cm hematomas; the 3 in the UFH group included three > 5 cm hematomas and 1 also required > 2 units packed cell transfusion. Minor vascular access complications for iliac PPIs in the bivalirudin group were all 50% restenosis requiring angiography and repeat PPI for bivalirudin and UFH patients, respectively. At the 6-month renal duplex ultrasound follow-up, 7/180 (3.88%) of bivalirudin-treated and 9/180 (5.0%) of UFH-treated patients developed > 50% restenosis requiring angiography and repeat PPI. Discussion Consistent with results in PCI utilizing bivalirudin, the analysis of the renal and iliac PPIs described above suggests that bivalirudin as the sole anticoagulant in PPI is safe, feasible and may have several advantages over UFH. Positive clinical outcomes favoring bivalirudin were comparable to previously observed experience with UFH and suggest potentially significant procedural advantages in shorter time to sheath removal, faster time to ambulation, and reduced LOS for patients treated with bivalirudin. LOS outcomes in renal PPI patients administered bivalirudin in this study suggest renal PPI may be suitable as an outpatient or 24 hours. A combination of several factors was considered leading to a decision to incorporate direct thrombin inhibition with bivalirudin to the anticoagulation strategy of treating our patients with PVD. These factors included: A) increased understanding and clinical awareness of UFH limitations; B) the favorable biochemical, pharmacokinetic and pharmacodynamic profile of the bivalirudin molecule; C) consistent reduction in ischemic and hemorrhagic complications reported in multiple PCI trials, especially the high-risk patient subsets; and D) the complex and distinct anatomical and clinical characteristics of PVD appear particularly well suited for direct thrombin inhibition. Heparin limitation relevant to PPI. UFH has been used extensively in PCI and PPI because of familiarity with the agent, because it is inexpensive and because attractive alternatives have been limited. Significant limitations in the pharmacokinetics, pharmacodynamics and clinical effectiveness of UFH have been identified (Table 5) and clearly reveal the need for better alternative anticoagulants. Prepared by crude extraction from either porcine intestine or bovine lung, UFH is a heterogeneous mixture of molecules varying with respect to molecular size, length and anticoagulation activity.10 UFH has non-linear pharmacokinetics, binds non-specifically to plasma proteins and endothelial cells, activates and increases platelet aggregation, fails to inhibit clot-bound thrombin and forms heparin antibodies that can increase the risk of heparin-induced thrombocytopenia (HIT) and thrombosis syndrome (HITTS).9 All these characteristics can adversely influence the clinical outcomes in PCI and PPI.9 Because of its unpredictable anticoagulant effect, patient-to-patient variable, and dose-dependent half-life, accurate dosing requires frequent monitoring of anticoagulant activity.9 The ACT is considered the “gold standard” for UFH monitoring, but with a variety of new ACT monitoring devices, different reagents and monitoring procedures, the relationship between ACT and anticoagulation status is not as clear-cut as previously thought.11 Of all the limitations associated with UFH, perhaps the 2 most significant are its inability to inhibit clot-bound thrombin12 and its formation of antibodies that can lead to HIT or HITTS.13 Thrombin is the main effector of the coagulation cascade and plays a pivotal role in arterial thrombosis. Thrombin acts as a potent platelet agonist and “procoagulant” by: 1) initiating activation and aggregation of platelets; 2) converting fibrinogen to fibrin, thus providing a cross-linking fibrin mesh stabilizing the thrombus; and 3) serving as a reservoir for further thrombin generation.14 Importantly, thrombin remains active even when bound to fibrin within a clot. The high association of thrombus in patients presenting with ACS and in PVD patients presenting with critical limb ischemia15 certainly raises questions regarding optimal efficacy of UFH as the anticoagulation foundation in these high-risk clinical scenarios. The HIT syndrome is a prothrombotic disorder characterized by intense activation of the coagulation system.13 The incidence of HIT is estimated at between 2–7% of patients exposed to UFH.16,17 With approximately 12 million patients exposed to UFH each year, nearly 360,000 patients are at risk for HIT.13 Of these (approximately 120,000), thirty percent are at risk for hemorrhagic, ischemic or thrombotic complications.13 A widespread vascular thrombosis can occur (HITTS) and can result in significant mortality (10–20%) and morbidity (10–20% amputation rate).17–19 Studies in patients with HIT suggest that 39–52% are at risk for a thrombotic complication within 30 days of diagnosis.16,20 Antibody formation occurs in approximately 61% of patients having cardiac surgery.21 In the post-cardiac surgery patients with positive antibodies, a 50% increase in thrombotic complications has been reported in the first year after surgery.22 Interestingly, 30–60% of the thrombotic events have been deep venous thrombosis.22 Certainly, anticoagulation strategies aimed at reducing HIT will become very important with our aging population and increasing number of cardiovascular and peripheral procedures. Novel non-UFH anticoagulation strategies designed to enhance the safety and clinical efficacy profile of PCI have shown reductions in both hemorrhagic and ischemic complications, raising the possibility of replacing UFH as the anticoagulation foundation in PCI.7,23,24 Unique mechanism of action and pharmacology of bivalirudin. Bivalirudin, a direct thrombin inhibitor, is a synthetic, small, 20-amino acid polypeptide with a molecular weight of 2,180 Daltons and a plasma half-life of 24 minutes.14,16 Unlike UFH, bivalirudin inhibits thrombin in both its plasma fluid and fibrin-bound phase, thus providing for a more complete inhibition of thrombin, as well as inhibition of thrombin-mediated platelet activation.12 By binding directly to thrombin without the need for the antithrombin cofactor required by UFH, bivalirudin is further differentiated from UFH. Bivalirudin binds directly to both the active site and exosite-1 (fibrin binding site) of the thrombin molecule in a reversible manner.14 Thrombin cleaves bivalirudin from the active site, leaving the remaining portion of the bivalirudin molecule bound to the exocite-1 with low affinity. A competitive inhibition environment results, allowing fibrin to displace bivalirudin. Once an infusion is discontinued, this competitive inhibition and the 25-minute half-life of bivalirudin allow thrombin to return to full hemostatic function within 1-2 hours.12,14 This mechanism is thought to be a key contributor to the favorable safety profile seen in clinical trials with bivalirudin. Bivalirudin is cleared from the plasma by proteolytic cleavage (80%) and renal clearance (20%) with elimination associated with the glomerular filtration rate (GFR).25 Pharmacokinetic data suggest that the bivalirudin half-life is prolonged from 24 minutes in patients with normal renal function to 34 minutes and 57 minutes in patients with moderate and severe renal impairment, respectively. Consequently, ACT monitoring is recommended in patients with chronic renal disease (CRD) and the bivalirudin dose reduced for patients with increasing severity of renal impairment.25,26 In further contrast to UFH, bivalirudin does not bind with other cellular or plasma proteins so is completely bioavailable (versus 30–35% with UFH), exhibits reliable linear dosing and concentration-respondent anticoagulation activity, negating the need for frequent ACT monitoring, and is not inhibited by PF-4.16,27,28 Bivalirudin is a small molecule with minimal secondary structure and is therefore unlikely to be immunogenic, is not associated with HIT or HITTS and is unlikely to cause thrombocytopenia.13,16 While the ability to reverse the anticoagulant effects of UFH with protamine sulfate may provide a degree of comfort when using UFH, significant hypotension and severe allergic reaction to protamine can be associated with significant morbidity.29,30 Consequently, complete avoidance of protamine is clinically desirable. There is no reversal agent for the anticoagulant effects of bivalirudin, which might call into question what to do in the rare case of perforation. Approximately 20% of the active drug is dialyzable. While fresh-frozen plasma (FFP) can be used as a management tool, it does not reverse the anticoagulant effects of bivalirudin. However, the short half-life (24 minutes) of bivalirudin coupled with the apparent rapid return to hemostasis due to the cleavage of bivalirudin from the thrombin molecule appears to have particular advantages in managing vascular access hemostasis in PPI patients. As a practical matter, potential perforations can be controlled by inflating a balloon and waiting 20–30 minutes for the bivalirudin serum levels to decrease. After 20–30 minutes, deflate the balloon and if no perforation is apparent, stent the lesion if not previously stented. A covered endograft is recommended if any concerns of perforation remain. Clinical outcomes with bivalirudin in PCI. Bivalirudin was associated with fewer ischemic (22%) and hemorrhagic complications (62%) in unstable angina patients undergoing angioplasty when compared to UFH alone in the Bivalirudin Angioplasty Trial (BAT).7 This trial also demonstrated significant reductions in complications in a high-risk post-MI population treated with bivalirudin compared to heparin. Although this trial pre-dated the routine use of stents, thienopyridines and GP IIb/IIIa inhibitors, the dual reduction of ischemic and hemorrhagic complications provided strong incentive to further evaluate this agent in a more contemporary interventional setting. The CACHET B/C and REPLACE-1 trials provided initial dosing and safety information on the use of bivalirudin in a modern cath lab setting.23,24 The unique dual reduction in ischemic and hemorrhagic complications was again evident in these exploratory trials (Table 6). Notably, in REPLACE-1 where GP IIb/IIIa inhibitor use was 72% for both treatment arms, reductions in ischemic and bleeding events were evident and provided the rationale for the recently published REPLACE-2 trial.24 Results from REPLACE-2 (n = 6002) have solidified the role of bivalirudin as an alternative anticoagulant to heparin in PCI. GP IIb/IIIa inhibitors were used in only 7.2% of the patients receiving bivalirudin. The primary quadruple endpoint of death, MI, urgent revascularization at 30 days and major in-hospital bleeding was not significantly different between groups (9.2% for bivalirudin with provisional GP IIb/IIIa use versus 10.0% for UFH and planned GP IIb/IIIa use), nor was there a significant difference between the triple endpoint of death, MI or revascularization (7.6% versus 7.1%).11 There was, however, a statistically significant reduction in major bleeding events in patients treated with bivalirudin (2.4% versus 4.1%; p = 0.001). Results from the REPLACE-2 trial provide substantial evidence that bivalirudin is a better foundation anticoagulant than UFH alone, while providing ischemic protection that is as effective as UFH with GP IIb/IIIa receptor antagonists but with significantly less bleeding.11 Collectively, these trials provided strong evidence of the utility of bivalirudin in improving patient outcomes in a contemporary setting employing stents, thienopyridines and GP IIb/IIIa inhibitors. The need to replace UFH in the treatment of PVD may be even more compelling. Predictors for post-PCI complications that are likely applicable to the PPI patient population. Multiple large PCI trials have consistently identified those patients at high risk for both hemorrhagic and ischemic complications. Potential predictors of post-PCI hemorrhagic and ischemic complications include female gender31,32 advanced age5,31,33 chronic renal disease34 acute ischemic event and diabetes.31,33 Larger sheath size is also an important predictor of increased bleeding in both PCI and PPI.4 Additional predictors of hemorrhagic and ischemic complications are listed in Table 7. Sparse data exist in the literature reporting or discussing hemorrhagic or ischemic complications or their risk factors in the post-PPI patient. Although this study is too small to assess meaningful correlations between risk factors and outcomes in PPI, it could easily be theorized that most of the predictors for post-PCI adverse outcomes would likewise be applicable to the PPI patient population. It may also be speculated that the benefits of more complete thrombin inhibition and reversible binding seen with bivalirudin in PCI as reduced ischemic and bleeding complications in high-risk patients compared to heparin would also be apparent in PPI and warrant further investigation. Ischemic endpoints may be less well defined in treating PVD, but subacute stent thrombosis does occur and in general data suggest that the PPI outcomes show worse acute and long-term results as compared to PCI outcomes, especially in treating lower extremity PVD.35,36 Bleeding complications have been described in treating patients with critical limb ischemia and thrombolysis,15,37,38 but few data exist regarding hemorrhagic complication rates in more elective PPI patients. Diabetes mellitus was present in approximately 20% of PCI trials, but has been reported at a much higher incidence (50%) in patients requiring lower extremity PPI.39,40 Diabetes mellitus has been associated with increased intracoronary thrombus formation due to enhanced plaque vulnerability and ulceration, leading to increased ischemic complications and restenosis post-PCI.33,41 A clear risk factor for post-PCI complications, there is no reason to suspect that diabetes mellitus is not also a significant risk factor for increased ischemic and bleeding complications in PPI. Chronic renal disease (CRD) has only recently been recognized as a predictor of adverse outcomes post-PCI with an increase in both hemorrhagic and ischemic outcomes.33,34 CRD is associated with increased thrombin generation, a pro-inflammatory atherogenic state, and complex platelet dysfunction.42 Even mild renal insufficiency is associated with poor outcomes.34,42 Similarly, CRD has long been recognized as a marker for complications and poor outcomes in the treatment of PVD.36,40 Complex extensive calcific PVD is associated with CRD, further complicating treatment, and limb salvage results in the past have been so dismal that primary amputation instead of revascularization has been recommended.38,43–45 The incidence of CRD in PCI trials has been estimated at 20%.34 It is highly likely that the incidence of CRD will be even higher in the PVD patient, especially in the older diabetic population presenting with lower extremity vascular insufficiency. The PVD patient with CRD is a particularly high-risk patient population needing optimal anticoagulation to maximize outcomes and minimize complications. In BAT, approximately 75% of patients had some degree of renal insufficiency.26 A retrospective analysis of the BAT trial was performed by renal function. As the severity of renal impairment increased, the rates of ischemic and hemorrhagic complications also increased. However, complications occurred less frequently regardless of renal function in the bivalirudin group versus the heparin group, even in the absence of any dose adjustments.4 While evidence suggests that UFH is associated with both increased ischemic and hemorrhagic complications in patients with CRD, there is no accepted recommendation for UFH dose adjustment in patients with renal impairment undergoing PCI.9 Based on the pharmacokinetic profile of bivalirudin, dose reductions of 20% and 60% are recommended in patients undergoing PCI with moderate (GRF 30–59 ml/minute) or severe renal (GFR 10–29 ml/minute) impairment, respectively. While it may be theorized that a similar recommendation for dose reduction could be made for treating patients with PVD and CRD, clinical data supporting such a recommendation need to be generated. Distinctive characteristics of PVD. Several clinical characteristics inherent to treating patients with PVD may make the PPI patient at even higher risk for hemorrhagic and ischemic complications than the PCI patients (Table 8). Larger sheath sizes and longer procedural duration times with PPI treatment compared to PCI, and the critical limb ischemia patient who will often require “thrombolytic therapy”, mechanical thrombectomy, or a combination of both, underscore the complexity of treating PVD. In theory, blood flow velocity is also much slower in the peripheral vessels than coronary vessels, especially in the superficial femoral artery. Basic principles of physics have established that given a constant pressure, the velocity of a fluid through a cylinder will be related to the diameter of that cylinder. The rate of flow will decrease as the cylinder becomes larger, or conversely, the velocity will increase as the diameter of the cylinder narrows. The driving force behind blood flow is greater in the coronaries that come off the aortic root than in peripheral vessels. In patients with PVD, this fall off of pressure can result in slower flow and potentially lead to increased thrombogenicity in PPI versus PCI and emphasizes the need for effective anticoagulation. These anatomic and technique-related differences in PPI versus PCI treatment may make treatment of PVD particularly prone to hemorrhagic, thrombotic and ischemic complications, again underscoring the need for an optimal anticoagulation strategy in treating PVD. The mechanism of action, pharmacokinetic profile and clinical outcomes with bivalirudin in PCI suggest that this agent may be equally beneficial in inhibiting clot-bound and soluble thrombin and thrombin-mediated platelet activation, thereby reducing the risk of thrombus formation and ischemic events while simultaneously reducing the risk of hemorrhagic complications in PPI. Study limitations. Inherent limitations of this study include the small sample size, a single center experience, use of a historical control for comparison and the potential observational study bias (selection bias) with the retrospective nonrandomized study design. Use of different types and lengths of stent over the time period covered in this study may also introduce bias. A detailed analysis of vascular closure device use and the role of ACTs was not consistently available and is a definite limitation of this study. Despite these study limitations, bivalirudin appears to be a safe and feasible non-UFH alternative anticoagulant for renal and iliac PPI and may offer decreased sheath removal time, time to ambulation and LOS. Conclusion. This study of the 180 renal and 75 iliac PPIs suggests that using bivalirudin as the sole anticoagulant in PPI appears to be safe, feasible and may have several advantages over UFH. There were no periprocedural deaths, MIs, intracranial bleeding, cerebrovascular accidents or major complications requiring surgery in either group. Fewer vascular access complications were observed in the bivalirudin versus the UFH group for both renal and iliac PPIs. Significant advantages may exist in shorter time to sheath removal, faster time to ambulation, and reduced LOS for patients treated with bivalirudin. The sheath removal time for renal PPIs was 24 hours. Several clinical characteristics of the PVD patient make bivalirudin particularly attractive as the anticoagulation foundation during PPI. A larger, prospective, randomized multicenter trial is warranted.
1. Krajcer Z, Howell MH. Update on endovascular treatment of peripheral vascular disease: New tools, techniques and indications. Tex Heart Inst J 2000;27:369–385. 2. Becker GJ, McClenny TE, Kovacs ME, et al. The importance of increasing public awareness of peripheral artery disease. J Vasc Interv Radiol 2002;13:7–11. 3. Gruentzig AR, Senning A, Siegenthaler WE. Nonoperative dilation of coronary artery stenosis. N Engl J Med 1979;301:61–68. 4. Aguirre FV, Gill JB. Increasing benefit, reducing risk: Focusing on hemorrhagic complications in percutaneous coronary intervention. J Invas Cardiol 2002;14:48B–54B. 5. The CAPTURE Investigators. Randomized placebo-controlled trial of abciximab before and during coronary intervention in refractory unstable angina: The CAPTURE study. Lancet 1997;349:1429–1435. 6. Topol EJ, Moliterno D, Herrman H, et al. Comparison of two platelet glycoprotein IIb/IIIa inhibitors, tirofiban and abciximab, for the prevention of ischemic events with percutaneous coronary revascularization. N Engl J Med 2001;344:1888–1894. 7. Bittl JA, Chaitman B, Feit F, et al. Bivalirudin versus heparin during coronary angioplasty for unstable or post-infarction angina: The final reanalysis of the bivalirudin angioplasty study. Am Heart J 2001;142:952–959. 8. Ouriel K. Peripheral arterial disease. Lancet 2001;358:1257–1264. 9. Hirsh J, Warkentin TE, Raschke R, et al. Heparin and low molecular weight heparin: Mechanisms of action, pharmacokinetics, dosing considerations, monitoring, efficacy and safety. Chest 1998;114:489S–510S. 10. Linhardt RJ, Gunay NS. Production and chemical processing of low molecular weight heparins. Semin Thromb Hemostasis 1999;25(Suppl 3):5–16. 11. Lincoff AM, Bittl JA, Harrington RA, et al., for the REPLACE-2 Investigators. Bivalirudin and provisional GP IIb/IIIa blockade compared with heparin and planned GP IIb/IIIa blockade during percutaneous coronary angioplasty. JAMA 2003;289:853–863. 12. Weitz JI, Bates SM. Acute coronary syndromes: A focus on thrombin. J Invas Cardiol 2002;14(Suppl B):2B–7B. 13. Lewis BE. Thrombocytopenia and outcomes in invasive cardiology. J Invas Cardiol 2002;14(Suppl B):38B–47B. 14. Bates SM, Weitz JI. Direct thrombin inhibitors for treatment of arterial thrombosis: Potential differences between bivalirudin and hirudin. Am J Cardiol 1998;82:12P–18P. 15. Semba CP, Murphy TP, Bakal CW. Thrombolytic therapy with use of alteplase (rt-PA) in peripheral arterial occlusive disease: Review of the clinical literature. J Vasc Interv Radiol 2000;11:149–161. 16. Campbell KR, Mahaffrey KW, Lewis BE, et al. Bivalirudin in patients with heparin-induced thrombocytopenia undergoing percutaneous coronary intervention. J Invas Cardiol 2000;12:14F–19F. 17. Warkentin TE. Clinical presentation of heparin-induced thrombocytopenia. Semin Hematology 1998;35(Suppl 5):9–16. 18. Kelton JG. The clinical management of heparin-induced thrombocytopenia. Semin Hematology 1999;36(Suppl 1):17–21. 19. Warkentin TE. Heparin-induced thrombocytopenia: Pathogenesis, frequency, avoidance and management. Drug Safety 197;5:325–341. 20. Warkentin TE, Kelton JG. A 14-year study of heparin-induced thrombocytopenia. Am J Med 1996;101:502–507. 21. Visentin GP, Malik M, Cyganiak KA, Aster. Patients treated with unfractionated heparin during open heart surgery are at high risk to form antibodies reactive with heparin: Platelet factor 4 complexes. J Lab Clin Med 1996;128:376–383. 22. Mattoili AV, Bonetti L, Sternieri S, Mattioli G. Heparin-induced thrombocytopenia in patients treated with unfractionated heparin: Prevalence of thrombosis in 1 year follow-up. Ital Heart J 2000;1:39–42. 23. Lincoff AM, Kleiman NS, Kottke-Marchant K, et al. Bivalirudin with planned or provisional abciximab versus low-dose heparin and abciximab during percutaneous coronary revascularization: Results of the Comparison of Abciximab Complications with Hirulog for Ischemic Events Trial (CACHET). Am Heart J 2002;143:847–853. 24. Lincoff AM. Pilot study of bivalirudin versus heparin during percutaneous coronary intervention with stenting and GP IIb/IIIa blockade: Results of the REPLACE-1 trial (Abstr). J Am Coll Cardiol 2002;39(Suppl A):16A. 25. Robson R, White H, Aylward P, Frampton C. Bivalirudin pharmacokinetics and pharmacodynamics: Effect of renal function, dose and gender. Clin Pharmacol Ther 2002;71:433–439. 26. Robson R. The use of bivalirudin in patients with renal impairment. J Invas Cardiol 2000;12(Suppl F):33F–36F. 27. Lui HK. Dosage, pharmacological effects and clinical outcomes for bivalirudin in percutaneous coronary intervention. J Invas Cardiol 2000;12:41F–52F. 28. Topol EJ, Bonan R, Jewitt D, et al. Use of direct thrombin inhibitor, hirulog, in place of heparin during coronary angioplasty. Circulation 1993;87:1622–1629. 29. Stewart WJ, McSweeney SM, Kellett MA, et al. Increased risk of severe protamine reactions in NPH insulin-dependent diabetic undergoing cardiac catheterization. Circulation 1984;70:788–792. 30. Vincent GM, Janowski M, Menlove R. Protamine allergy reactions during cardiac catheterization and cardiac surgery: Risk in patients taking protamine-insulin preparation. Cathet Cardiovasc Diagn 1987;13:214–217. 31. Aguirre FV, Topol EJ, Ferguson JJ, et al. Bleeding complications with the chimeric antibody to platelet glycoprotein IIb/IIIa integrin in patients undergoing percutaneous coronary intervention. The EPIC Investigators. Circulation 1995;91:2882–2890. 32. Ellis SG. Elective coronary angioplasty: Techniques and complications. In: Topol E (ed). Textbook of Interventional Cardiology. WB Saunders: Philadelphia, 1999: pp. 147–162. 33. Lincoff AM, Topol E. Abrupt vessel closure. In: Topol E (ed). Textbook of Interventional Cardiology. WB Saunders: Philadelphia, 1999: pp. 163–187. 34. Best P, Lennon R, Ting H, et al. Even mild renal insufficiency is associated with increased mortality after percutaneous coronary intervention. J Am Coll Cardiol 2001;37:76A. 35. Capek P, McLean GK, Berkowitz HD. Femoropopliteal angioplasty. Factors influencing long-term success. Circulation 1991;83(Suppl 2):170–180. 36. Hunink MRM, Magruder CD, Meyerovitz MR, et al. Risks and benefits of femoropopliteal percutaneous balloon angioplasty. J Vasc Surg 1993;17:183–194. 37. Ouriel K, Shortell CK, DeWeese JA, et al. A comparison of thrombolytic therapy with operative revascularization in the initial treatment of acute peripheral arterial ischemia. J Vasc Surg 1994;19:1021–1030. 38. Ouriel K. Surgery versus thrombolyitc therapy in the management of peripheral arterial occlusions. J Vasc Interv Radiol 1995;6:48S–54S. 39. Miller JM, Ohman EM, Moliterno DJ, et al. Restenosis: The clinical issues. In: Topol E (ed). Textbook of Interventional Cardiology. WB Saunders: Philadelphia, 1999: pp. 379–415. 40. Henry TD. Overcoming heparin limitations in high-risk percutaneous coronary intervention: The alternative strategy — Replacing heparin with bivalirudin. J Invas Cardiol 2002;14:19B–26B. 41. Rosenfield K, Schainfeld RS, Isnor JM. Percutaneous revascularization in peripheral arterial disease. Current Prob Cardiol 1996;31:1–96. 42. Asinger RW, Henry HD, Herzog CA, et al. Clinical outcomes of PTCA in chronic renal failure: A case-control study for comorbid features and evaluation of dialysis dependence. J Invas Cardiol 2001;13:21–28. 43. Gray BH, Sullivan TM, Childs MB, et al. High incidence of restenosis/reocclusion of stents in the percutaneous treatment of long segment superficial femoral artery disease after suboptimal angioplasty. J Vasc Surg 1997;25:74–83. 44. Ouriel K, Shortell CK, Azodo MVU, et al. Acute peripheral vascular occlusion: Predictors of success in catheter-directed thrombolytic therapy. Radiology 1994;193:561–566. 45. Ouriel K, Veith FJ, Sarahara AA. A comparison of recombinant urokinase with vascular surgery as initial treatment for acute arterial occlusion of the legs. N Engl J Med 1998;338:1105–1111. 46. Xiao Z, Theroux P. Platelet activation with unfractionated heparin at therapeutic concentrations and comparisons with a low molecular weight heparin and with a direct thrombin inhibitor. Circulation 1998;97:251–256