American Journal of Therapeutics 21, e100–e105 (2014)
Extensive Fatal Intracoronary Thrombosis During Percutaneous Coronary Intervention With Bivalirudin Sanjiv Sharma, MD, FACC, FSCAI,1,2* Shirish Patel, MD, FACC,3 Ashok Behl, MD, FACC,1 Sarabjeet Singh, MD,1,2 Rasham Sandhu, MD,1,2 Neil Bhambi, BA,2 Rohan Sharma,2 and Brijesh Bhambi, MD, FACC, FSCAI1,2
The authors describe 2 cases of extensive intracoronary thrombus formation leading to acute closure of the left main where bivalirudin (Angiomax) was used as the anticoagulant during percutaneous coronary intervention leading to mortality. Both cases had similarity in the cascade of complications of coronary dissection leading to slow flow and prolonged procedure time with compromise of antegrade flow in the coronary artery and a final catastrophic development of extensive intracoronary thrombosis extending into the left main and nonintervened vessel (left anterior descending or circumflex) followed by ventricular fibrillation and death. Bivalirudin has reversible anticoagulant pharmacodynamics because the bivalirudin molecule is cleaved by the thrombin molecule. In situations when the antegrade flow is compromised, delivery of fresh circulating bivalirudin to replenish the catalysis of bivalirudin by thrombin is diminished, allowing thrombin activity to regenerate, thereby creating a prothrombotic milieu in these coronary segments. This can lead to extensive intracoronary thrombus formation in situations of slow flow precipitated by coronary dissection and prolonged dwell time with intracoronary hardware (wires, balloons, and stents). Interventionalists should be aware of the potential risk of this fatal complication and should be proactive in recognizing the scenarios where this is likely to occur. In such anticipated circumstances, the interventionalist may judiciously switch the anticoagulant to heparin and/or use additional glycoprotein IIb/IIIa inhibitor because freshly formed intracoronary thrombus is susceptible to lysis by glycoprotein IIb/IIIa inhibitors. Keywords: bivalirudin, coronary thrombus, angioplasty
INTRODUCTION Bivalirudin (Angiomax; The Medicines Co, Parsippany, NJ) had been used as the anticoagulant for the percutaneous coronary intervention (PCI) procedures. The use of the drug has increased after publication of data from the Acute Catheterization and Urgent Intervention
1
Cardiology Division, Department of Medicine, Bakersfield Heart Hospital, Bakersfield, CA; 2Central Cardiology Medical Clinic, Bakersfield, CA; and 3Community Memorial Hospital, Ventura, CA. The authors have no conflicts of interest to declare. *Address for correspondence: Bakersfield Heart Hospital, 2901 Sillect Avenue, Ste 100, Bakersfield, CA 93308. E-mail: [email protected] 1075–2765 Ó 2013 Lippincott Williams & Wilkins
Triage strategy (ACUITY) and Horizons-AMI trials.1,2 The drug has been cited as efficacious, although concerns have been raised about a higher incidence of acute stent thrombosis in the Horizons-AMI trial. Most of the data about the drug relate to the equivalency or superiority in terms of combined end points of death, myocardial infarction (MI), and bleeding. The lower bleeding events after use of bivalirudin are because of the reversible anticoagulant effect of the drug related to its reversible kinetics. Thrombin inhibition effected by the drug is reversible as the thrombin molecule cleaves the bivalirudin molecule bound to itself and then the thrombin activity returns.3 We postulate that this return of thrombin activity is predisposed to occur in states of slow flow and stasis where thrombin has time to recover its inhibition by bivalirudin because replenishment of the fresh www.americantherapeutics.com
Intracoronary Thrombosis With Bivalirudin
bivalirudin from the circulating drug does not occur and the thrombin–bivalirudin complexes in the segment of circulation with stasis essentially follows the slow degradation of bivalirudin by thrombin and eventual recovery of thrombin activity.4 The mechanism of the acute extensive intracoronary thrombus formation is likely related to the reversible anticoagulant effect that follows degradation of bivalirudin and return of thrombin procoagulant activity.
CASE REPORTS Case 1 The patient was a 45-year-old man with type 2 diabetes mellitus and PCI and stenting of the left anterior descending (LAD) in the past. He developed symptoms of recurrent angina prompting angiography. The patient was noted to have a focal-edge restenosis of the LAD stent. The patient underwent PCI of the LAD in-stent restenosis lesion (Figure 1). He had been on aspirin therapy and he received a dose of aspirin 325 mg on the day of the procedure. Bivalirudin (Angiomax; The Medicines Co), a direct thrombin inhibitor, was used as the anticoagulant for the PCI procedure with a 1 mg/kg intravenous bolus administered before the start of the PCI followed by 2.5 mg21$kg21$h21 intravenous infusion during the procedure. An activated
FIGURE 1. Preintervention angiogram of case 1 showing focal in-stent restenosis lesion in the LAD artery. www.americantherapeutics.com
e101
clotting time of .400 seconds was obtained with the Hemochron system. The lesion in the LAD was crossed without difficulty using a 0.01499 175-cm guide wire. Predilatation of the lesion was carried out using a 3.5-mm cutting balloon (Flextome; Boston Scientific, Natick, MA) at 6–8 atmospheres. There was residual stenosis at the site of the lesion so additional dilations were carried out with an Apex 3.5-mm balloon (Boston Scientific) at nominal pressures (10–12 atmospheres). The segment of injury was documented under cinefluoroscopy. Angiography showed an area of haziness with the appearance of threatened abrupt closure at the distal edge of the treatment zone (distal to the stent). Multiple doses of intracoronary nitroglycerin (200 mg) and verapamil (200 mg) did not lead to resolution of the area of haziness, slow flow, and acute closure. Finally, there was extensive intracoronary thrombus noted in the left main extending into the LAD and the nonintervened left circumflex (Figure 2) with slow flow in the left main complicated with ventricular fibrillation cardiac arrest and death, despite resuscitation attempts. Case 2 The patient was a 66-year-old man who was referred for coronary angiography for symptoms of angina and a positive stress echocardiogram. Angiography showed a significant lesion in the left circumflex obtuse marginal branch bifurcation (Figure 3). Angiomax was
FIGURE 2. Angiogram showing extensive intracoronary thrombus formation in the left main, LAD, and nonintervened left circumflex, associated with slow flow in case 1. American Journal of Therapeutics (2014) 21(4)
e102
administered and continued for anticoagulation during the procedure. The left circumflex obtuse marginal branch lesion was crossed with a Choice PT guide wire (Boston Scientific). The PCI procedure was a complex procedure during which the vessel developed coronary dissection, slow flow, and intracoronary thrombus. The tip of the coronary guide wire broke and embolized into the target vessel. Stenting of the lesion was successfully accomplished. Attempts to retrieve the embolized wire with a aspiration catheter and finally an attempt to stent the embolized tip of the guide wire against the vessel wall were not successful because of the difficulty in delivering the aspiration catheter or the second stent past the previously deployed stent. During the procedure, acute threatened closure of the vessel started to occur. The procedure was prolonged with inability in restoring coronary flow. Eventually, there was development of extensive intracoronary thrombus noted, even in the nonculprit vessels (LAD and left main), associated with slow flow in the left coronary circulation (Figure 4). Emergent stenting of the left main into the LAD was accomplished and an intraaortic balloon pump was placed. However, meaningful flow could not be restored in the left main. The patient continued to suffer hemodynamic instability and recurrent episodes of ventricular fibrillation and could not be resuscitated from cardiac arrest, despite prolonged cardiopulmonary resuscitation efforts.
FIGURE 3. Preintervention angiogram of the left circumflex lesion in case 2. American Journal of Therapeutics (2014) 21(4)
Sharma et al
FIGURE 4. Angiogram showing extensive intracoronary thrombus formation in the left main, left circumflex, and nonintervened LAD, associated with slow flow in case 2.
DISCUSSION There is scant literature about the development of coronary thrombus after use of bivalirudin during routine PCI. The package insert of Angiomax lists “thrombus formation during PCI with or without intracoronary brachytherapy including reports of fatal outcomes” can occur with the use of the drug. However, awareness and understanding of the settings in which thrombus formation during PCI with use of bivalirudin occurs is not widely known. Thrombin plays a central role in thrombus production at sites of arterial injury during PCI. Thrombin acts by cleaving fibrinogen to fibrin and activating factor XIII leading to thrombus formation and by activating platelets causing platelet aggregation and release of platelet granules (mediators). Thrombin also activates factors V and VIII thereby causing further thrombin production and causing an amplification of thrombin activity.3 Bivalirudin (Angiomax) is a synthetic 20-amino acid polypeptide that acts as a specific and reversible direct thrombin inhibitor, binding in a 1:1 molar ratio to both circulating and clot-bound thrombin.3 Bivalirudin exhibits a reliable and immediate dose-dependent anticoagulant effect with prolongation of the coagulation times, prothrombin time, activated partial thromboplastin time, thrombin time, and activated clotting time. The www.americantherapeutics.com
Intracoronary Thrombosis With Bivalirudin
biological half-life is approximately 25 minutes, and coagulation times return to baseline after approximately 1 hour after cessation of bivalirudin administration. A single bivalirudin molecule binds in a 1:1 stoichiometric bivalent fashion to a single thrombin molecule at 2 sites: (1) carboxy-terminal segment of bivalirudin binds to exosite 1 of thrombin, which is also the binding site for fibrinogen and (2) the amino terminal of bivalirudin binds to the active (catalytic) site of thrombin, responsible for catalytic conversion of fibrinogen to fibrin and the activation of clotting factors and platelets. After this initial high-affinity binding of bivalirudin to thrombin, there is an initial complete inhibition of both the fibrinogen-binding and catalytic functions of thrombin. However, thrombin molecule slowly starts an enzymatic cleavage of bivalirudin near the amino-terminal end resulting in release of aminoterminal segment from the catalytic site, leaving the catalytic site free to perform catalytic functions like activation of platelets, besides fibrin production (Figure 5). As a result of the partial cleavage of bivalirudin by thrombin, the binding of the carboxy-terminal fragment of bivalirudin molecule to exosite 1 of thrombin becomes low affinity and weakly competitive. Fibrinogen molecule is able to now displace this weakly bound bivalirudin fragment from the exosite 1 (fibrinogen-binding site) of thrombin and thereby able to present itself to the catalytic site of thrombin to be converted to fibrin. In summary, the initial binding of bivalirudin to thrombin causes a complete but temporary inhibition of thrombin. This is followed by slow cleavage of bivalirudin, leading to a slow recovery of thrombin function.5 The binding of bivalirudin to thrombin in a 1:1 ratio implies that each molecule of bivalirudin can only inhibit the action of a single molecule of thrombin (unlike heparin, which can free itself and become available for potentiating additional molecules of antithrombin). The gradual enzymatic cleavage of bivalirudin by thrombin thereby requires the need for a continuous infusion of bivalirudin to replenish the antithrombin activity of the drug to ensure therapeutic level of anticoagulation.6 In the segment of the vessel with stasis or slow flow, there is no availability of a fresh supply of bivalirudin molecules (that are unable to reach the local milieu distally) and thrombin molecules slowly recover from inhibition, leading to development of a prothrombotic milieu in the vasculature. Thrombin then initiates a thrombotic cascade that conceivably propagates proximally as thrombin amplifies its own production (overtaking the anticoagulant effect of the bivalirudin) leading to extensive coronary thrombosis. These situations are likely to occur during flow-limiting coronary www.americantherapeutics.com
e103
FIGURE 5. (A) Initial binding of bivalirudin to thrombin occurs at the exosite 1 and active (catalytic) site of thrombin molecule, causing complete inhibition of thrombin. (B) Thrombin slowly cleaves bivalirudin causing the release of the amino-terminal fragment of bivalirudin from the active site, allowing partial recovery of thrombin catalytic function (like platelet activation, besides fibrin production). (C) Finally, the fibrinogen molecule displaces the weakly bound bivalirudin from thrombin, allowing full recovery of thrombin function (including fibrin production).
dissection, placement of coronary hardware limiting coronary flow for a period of time (prolonged balloon inflations, difficult delivery of stents, bulky devices like brachytherapy catheters, etc.), and slow flow or no-reflow phenomenon. Experimental data support the hypothesis that the anticoagulant effect of bivalirudin is overcome rapidly American Journal of Therapeutics (2014) 21(4)
e104
in the situations of stasis, when there is no replenishment of the drug by infusion. An in vitro study compared the development of catheter thrombosis for bivalirudin, enoxaparin, and unfractionated heparin (UFH) in a controlled in vitro environment.7 Ten healthy male volunteers were pretreated with aspirin 500 mg 2 hours before venesection of 50 mL of blood. The 7 groups of anticoagulant combinations tested were as follows: UFH, UFH + eptifibatide, enoxaparin, enoxaparin + eptifibatide, bivalirudin bolus, bivalirudin + eptifibatide, and bivalirudin bolus + continuous infusion. The blood/anticoagulant mixture was continuously circulated through a cardiac guiding catheter for 60 minutes or until the catheter became blocked with thrombus. After anticoagulation with bolus dose bivalirudin, the catheter was invariably occluded with thrombus after 33 minutes of circulation. However, a continuous infusion of bivalirudin prevented the development of occlusive catheter thrombosis. In the bolus bivalirudin group, the mean thrombus weight was significantly greater than in all other groups (P , 0.01 in all analyses). Bivalirudin given as a bolus was not sufficient to prevent cardiac catheter thrombosis in this in vitro study. However, a continuous infusion of bivalirudin had similar antithrombotic efficacy compared with other treatment strategies. Similarly, there is a case report of a large thrombus forming in the venous reservoir while using bivalirudin during cardiopulmonary bypass in a patient with heparin-induced thrombocytopenia. This was presumably related to stasis of blood associated with the full venous reservoir maintained in the case leading to formation of a large thrombus at the top of the venous canister.8 In the REPLACE-2 trial,9 the investigators allowed addition of glycoprotein IIb/IIIa inhibitors (GPI) to patients randomized to bivalirudin, on a “provisional” basis, in the following circumstances where there was a likely perceived benefit with the use of the GPI: decreased thrombolysis in myocardial infarction flow (0–2) or slow reflow, dissection with decreased flow, new or suspected thrombus, persistent residual stenosis, distal embolization, unplanned stent, suboptimal stenting, side branch closure, abrupt closure, clinical instability, and prolonged ischemia. During the study, 1 or more of these circumstances occurred in 12.7% of the patients in the bivalirudin (with provisional GPI) arm. The GPIs were administered to 7.2% of the patients in the bivalirudin (with provisional GPIs) arm (62.2% of the eligible patients). The composite triple end point of death, MI, or urgent revascularization was higher (7.6%) in the bivalirudin with “provisional” GPI arm as compared with the heparin plus American Journal of Therapeutics (2014) 21(4)
Sharma et al
GPI arm (7.1%). If these “bailout” provisional utilization of GPIs had not been allowed, it is conceivable that the composite triple end point of death, MI, or urgent revascularization could have been much higher in the bivalirudin group, as studies have shown that intracoronary thrombosis in the setting of bivalirudin use is reversible with the use of intracoronary abciximab.4 Whether use of bivalirudin imparts a higher risk of acute thrombosis in certain settings needs more conclusive data, although in some trials, there are suggestions of an effect pointing to this direction. In the HORIZONS-AMI trial,2 acute stent thrombosis occurred more frequently in patients assigned to bivalirudin compared with heparin plus a GPI (1.4% vs. 0.3%, P , 0.001). In the ACUITY trial,1 comparing UFH plus GPIs, bivalirudin plus GPI, and bivalirudin alone, patients treated with bivalirudin alone had an increase in ischemic events, especially if they were troponin positive and were not pretreated with a thienopyridine. A review of the literature on the frequency of coronary thrombosis occurring in the setting of PCI with bivalirudin reveals case reports in the Web sites of drug safety and side effects. Many authors have reported anecdotal cases of the coronary thrombus formation during PCI with bivalirudin.4,10,11 Kuchulakanti et al report 2 cases of intracoronary thrombosis with gamma radiation when bivalirudin is used as an anticoagulant. The authors used an experimental swine model to replicate their clinical observations. There was occurrence of thrombus in the guide catheter and over the intervention guide wire in 2 of the swine in the bivalirudin group. They postulated that thrombus forms because of prolonged dwell time and stasis of blood in the catheter and may propagate into the coronary arteries in some cases.6 In the Brachytherapy and Bivalirudin Evaluation Study,12 of the 152 patients who underwent PCI and vascular brachytherapy with either gamma or beta radiation, there was a high prevalence of acute procedural intracoronary thrombosis in patients treated with gamma irradiation (25%). Although the incidence of this complication was much lower (0.7%, P , 0.001) in the beta-irradiation group, acute procedural intracoronary thrombosis did still occur with use of bivalirudin. Sharma et al reported occurrence of acute closure because of intracoronary thrombus formation during vascular brachytherapy with beta radiation with use of bivalirudin in 2 patients.4 Upon removal of the Novoste brachytherapy catheter after a dwell time of approximately 3 minutes, there was acute closure of the vessel distal to the treatment zone. The acute closure was reversed with use of intracoronary administration of abciximab, signifying the presence of fresh platelet www.americantherapeutics.com
Intracoronary Thrombosis With Bivalirudin
thrombus that was amenable to rapid dissolution with delivery of high local dose (with intracoronary administration) of abciximab. The reversible pharmacodynamics of bivalirudin demands an understanding on part of the interventionalist that a recovery of thrombin function can follow the catabolism and unbinding of bivalirudin from thrombin. This can occur rapidly when there is absence of a replenishing delivery of the drug because of stasis or slow flow. This can tip the local milieu from an antithrombotic state toward a prothrombotic state locally at the intervention site and lead to development of extensive intracoronary thrombus formation with a high risk of mortality, as occurred in these 2 cases. We believe that prolonged PCI procedures associated with a situation of decreased TIMI flow (0–2) or slow reflow, flow-limiting dissection, new or suspected thrombus, distal embolization, and prolonged dwell times with any intervention hardware, especially bulky devices like brachytherapy catheter, can predispose to formation of profuse coronary thrombus with use of bivalirudin during PCI, with a risk of death. It demands an immediate treatment strategy with possibly a switch to heparin and a judicious use of GPIs in these situations.
CONCLUSIONS Extensive intracoronary thrombus formation with a high risk of mortality can follow prolonged PCI procedures associated with a situation of stasis and decreased TIMI flow (0–2) or slow reflow, flow-limiting dissection, and prolonged dwell times with any intervention hardware with use of bivalirudin because of its reversible kinetics of inhibition of thrombin. It demands an immediate recognition by the interventionalist, with immediate treatment strategy with possibly a switch to heparin and a judicious use of GPIs.
ACKNOWLEDGMENTS The authors would like to express their thanks to Mark Monterroso and Rohan Sharma for their assistance in preparation of the manuscript and figures. The authors also thank Dr Rasham Sandhu and Dr Brijesh Bhambi for their critical input in formulation of the final manuscript.
www.americantherapeutics.com
e105
REFERENCES 1. Stone GW, White HD, Ohman EM, et al; Acute Catheterization and Urgent Intervention Triage strategy (ACUITY) trial investigators. Bivalirudin in patients with acute coronary syndromes undergoing percutaneous coronary intervention: a subgroup analysis from the Acute Catheterization and Urgent Intervention Triage strategy (ACUITY) trial. Lancet. 2007;369:907–919. 2. Stone GW, Witzenbichler B, Guagliumi G, et al; HORIZONS-AMI Trial Investigators. Bivalirudin during primary PCI in acute myocardial infarction. N Engl J Med. 2008;358:2218–2230. 3. Angiomax (bivalirudin). Package Insert. Parsippany, NJ: The Medicines Company; 2010. 4. Sharma S, Bhambi B, Nyitray W, et al. Bivalirudin (Angiomax) use during intracoronary brachytherapy may predispose to acute closure. J Cardiovasc Pharmacol Ther. 2003;8:9–15. 5. Sciulli TM, Mauro VF. Pharmacology and clinical use of bivalirudin. Ann Pharmacother. 2002;36:1028–1041. 6. Kuchulakanti PK, Satler LF, Rha SW, et al. Bivalirudinassociated intracoronary thrombosis during gammabrachytherapy and its experimental validation in acute swine model. Catheter Cardiovasc Interv. 2004;62:209–213. 7. Maegdefessel L, Buerke M, Schubert S, et al. Comparison of bivalirudin, enoxaparin, and unfractionated heparin in preventing cardiac catheter thrombosis. Results of an in-vitro study. Thromb Haemost. 2008;100:693–698. 8. Wong JK, Tian Y, Shuttleworth P, et al. Case report: a thrombus in the venous reservoir while using bivalirudin in a patient with heparin-induced thrombocytopenia undergoing heart transplantation. Anesth Analg. 2010; 111:609–612. 9. Lincoff AM, Bittl JA, Harrington RA, et al; REPLACE-2 Investigators. Bivalirudin and provisional glycoprotein IIb/IIIa blockade compared with heparin and planned glycoprotein IIb/IIIa blockade during percutaneous coronary intervention: REPLACE-2 randomized trial. JAMA. 2003;289:853–863. 10. Tadros GM, Broder K, Bachour F. Intracoronary macrothrombus formation during percutaneous coronary intervention despite optimal activated clotting time using bivalirudin—a case report. Angiology. 2005;56: 761–765. 11. Chieffo A, Melzi G, Rogacka R, et al. Safety and feasibility of Bivalirudin with either Cypher and Taxus drugeluting stent during percutaneous coronary intervention. EuroIntervention. 2005;1:70–74. 12. Kuchulakanti P, Wolfram R, Torguson R, et al. Brachytherapy and bivalirudin evaluation study. Am Heart J. 2005;150:832–837.
American Journal of Therapeutics (2014) 21(4)
Extensive Fatal Intracoronary Thrombosis During Percutaneous Coronary Intervention With Bivalirudin Sanjiv Sharma, MD, FACC, FSCAI,1,2* Shirish Patel, MD, FACC,3 Ashok Behl, MD, FACC,1 Sarabjeet Singh, MD,1,2 Rasham Sandhu, MD,1,2 Neil Bhambi, BA,2 Rohan Sharma,2 and Brijesh Bhambi, MD, FACC, FSCAI1,2
The authors describe 2 cases of extensive intracoronary thrombus formation leading to acute closure of the left main where bivalirudin (Angiomax) was used as the anticoagulant during percutaneous coronary intervention leading to mortality. Both cases had similarity in the cascade of complications of coronary dissection leading to slow flow and prolonged procedure time with compromise of antegrade flow in the coronary artery and a final catastrophic development of extensive intracoronary thrombosis extending into the left main and nonintervened vessel (left anterior descending or circumflex) followed by ventricular fibrillation and death. Bivalirudin has reversible anticoagulant pharmacodynamics because the bivalirudin molecule is cleaved by the thrombin molecule. In situations when the antegrade flow is compromised, delivery of fresh circulating bivalirudin to replenish the catalysis of bivalirudin by thrombin is diminished, allowing thrombin activity to regenerate, thereby creating a prothrombotic milieu in these coronary segments. This can lead to extensive intracoronary thrombus formation in situations of slow flow precipitated by coronary dissection and prolonged dwell time with intracoronary hardware (wires, balloons, and stents). Interventionalists should be aware of the potential risk of this fatal complication and should be proactive in recognizing the scenarios where this is likely to occur. In such anticipated circumstances, the interventionalist may judiciously switch the anticoagulant to heparin and/or use additional glycoprotein IIb/IIIa inhibitor because freshly formed intracoronary thrombus is susceptible to lysis by glycoprotein IIb/IIIa inhibitors. Keywords: bivalirudin, coronary thrombus, angioplasty
INTRODUCTION Bivalirudin (Angiomax; The Medicines Co, Parsippany, NJ) had been used as the anticoagulant for the percutaneous coronary intervention (PCI) procedures. The use of the drug has increased after publication of data from the Acute Catheterization and Urgent Intervention
1
Cardiology Division, Department of Medicine, Bakersfield Heart Hospital, Bakersfield, CA; 2Central Cardiology Medical Clinic, Bakersfield, CA; and 3Community Memorial Hospital, Ventura, CA. The authors have no conflicts of interest to declare. *Address for correspondence: Bakersfield Heart Hospital, 2901 Sillect Avenue, Ste 100, Bakersfield, CA 93308. E-mail: [email protected] 1075–2765 Ó 2013 Lippincott Williams & Wilkins
Triage strategy (ACUITY) and Horizons-AMI trials.1,2 The drug has been cited as efficacious, although concerns have been raised about a higher incidence of acute stent thrombosis in the Horizons-AMI trial. Most of the data about the drug relate to the equivalency or superiority in terms of combined end points of death, myocardial infarction (MI), and bleeding. The lower bleeding events after use of bivalirudin are because of the reversible anticoagulant effect of the drug related to its reversible kinetics. Thrombin inhibition effected by the drug is reversible as the thrombin molecule cleaves the bivalirudin molecule bound to itself and then the thrombin activity returns.3 We postulate that this return of thrombin activity is predisposed to occur in states of slow flow and stasis where thrombin has time to recover its inhibition by bivalirudin because replenishment of the fresh www.americantherapeutics.com
Intracoronary Thrombosis With Bivalirudin
bivalirudin from the circulating drug does not occur and the thrombin–bivalirudin complexes in the segment of circulation with stasis essentially follows the slow degradation of bivalirudin by thrombin and eventual recovery of thrombin activity.4 The mechanism of the acute extensive intracoronary thrombus formation is likely related to the reversible anticoagulant effect that follows degradation of bivalirudin and return of thrombin procoagulant activity.
CASE REPORTS Case 1 The patient was a 45-year-old man with type 2 diabetes mellitus and PCI and stenting of the left anterior descending (LAD) in the past. He developed symptoms of recurrent angina prompting angiography. The patient was noted to have a focal-edge restenosis of the LAD stent. The patient underwent PCI of the LAD in-stent restenosis lesion (Figure 1). He had been on aspirin therapy and he received a dose of aspirin 325 mg on the day of the procedure. Bivalirudin (Angiomax; The Medicines Co), a direct thrombin inhibitor, was used as the anticoagulant for the PCI procedure with a 1 mg/kg intravenous bolus administered before the start of the PCI followed by 2.5 mg21$kg21$h21 intravenous infusion during the procedure. An activated
FIGURE 1. Preintervention angiogram of case 1 showing focal in-stent restenosis lesion in the LAD artery. www.americantherapeutics.com
e101
clotting time of .400 seconds was obtained with the Hemochron system. The lesion in the LAD was crossed without difficulty using a 0.01499 175-cm guide wire. Predilatation of the lesion was carried out using a 3.5-mm cutting balloon (Flextome; Boston Scientific, Natick, MA) at 6–8 atmospheres. There was residual stenosis at the site of the lesion so additional dilations were carried out with an Apex 3.5-mm balloon (Boston Scientific) at nominal pressures (10–12 atmospheres). The segment of injury was documented under cinefluoroscopy. Angiography showed an area of haziness with the appearance of threatened abrupt closure at the distal edge of the treatment zone (distal to the stent). Multiple doses of intracoronary nitroglycerin (200 mg) and verapamil (200 mg) did not lead to resolution of the area of haziness, slow flow, and acute closure. Finally, there was extensive intracoronary thrombus noted in the left main extending into the LAD and the nonintervened left circumflex (Figure 2) with slow flow in the left main complicated with ventricular fibrillation cardiac arrest and death, despite resuscitation attempts. Case 2 The patient was a 66-year-old man who was referred for coronary angiography for symptoms of angina and a positive stress echocardiogram. Angiography showed a significant lesion in the left circumflex obtuse marginal branch bifurcation (Figure 3). Angiomax was
FIGURE 2. Angiogram showing extensive intracoronary thrombus formation in the left main, LAD, and nonintervened left circumflex, associated with slow flow in case 1. American Journal of Therapeutics (2014) 21(4)
e102
administered and continued for anticoagulation during the procedure. The left circumflex obtuse marginal branch lesion was crossed with a Choice PT guide wire (Boston Scientific). The PCI procedure was a complex procedure during which the vessel developed coronary dissection, slow flow, and intracoronary thrombus. The tip of the coronary guide wire broke and embolized into the target vessel. Stenting of the lesion was successfully accomplished. Attempts to retrieve the embolized wire with a aspiration catheter and finally an attempt to stent the embolized tip of the guide wire against the vessel wall were not successful because of the difficulty in delivering the aspiration catheter or the second stent past the previously deployed stent. During the procedure, acute threatened closure of the vessel started to occur. The procedure was prolonged with inability in restoring coronary flow. Eventually, there was development of extensive intracoronary thrombus noted, even in the nonculprit vessels (LAD and left main), associated with slow flow in the left coronary circulation (Figure 4). Emergent stenting of the left main into the LAD was accomplished and an intraaortic balloon pump was placed. However, meaningful flow could not be restored in the left main. The patient continued to suffer hemodynamic instability and recurrent episodes of ventricular fibrillation and could not be resuscitated from cardiac arrest, despite prolonged cardiopulmonary resuscitation efforts.
FIGURE 3. Preintervention angiogram of the left circumflex lesion in case 2. American Journal of Therapeutics (2014) 21(4)
Sharma et al
FIGURE 4. Angiogram showing extensive intracoronary thrombus formation in the left main, left circumflex, and nonintervened LAD, associated with slow flow in case 2.
DISCUSSION There is scant literature about the development of coronary thrombus after use of bivalirudin during routine PCI. The package insert of Angiomax lists “thrombus formation during PCI with or without intracoronary brachytherapy including reports of fatal outcomes” can occur with the use of the drug. However, awareness and understanding of the settings in which thrombus formation during PCI with use of bivalirudin occurs is not widely known. Thrombin plays a central role in thrombus production at sites of arterial injury during PCI. Thrombin acts by cleaving fibrinogen to fibrin and activating factor XIII leading to thrombus formation and by activating platelets causing platelet aggregation and release of platelet granules (mediators). Thrombin also activates factors V and VIII thereby causing further thrombin production and causing an amplification of thrombin activity.3 Bivalirudin (Angiomax) is a synthetic 20-amino acid polypeptide that acts as a specific and reversible direct thrombin inhibitor, binding in a 1:1 molar ratio to both circulating and clot-bound thrombin.3 Bivalirudin exhibits a reliable and immediate dose-dependent anticoagulant effect with prolongation of the coagulation times, prothrombin time, activated partial thromboplastin time, thrombin time, and activated clotting time. The www.americantherapeutics.com
Intracoronary Thrombosis With Bivalirudin
biological half-life is approximately 25 minutes, and coagulation times return to baseline after approximately 1 hour after cessation of bivalirudin administration. A single bivalirudin molecule binds in a 1:1 stoichiometric bivalent fashion to a single thrombin molecule at 2 sites: (1) carboxy-terminal segment of bivalirudin binds to exosite 1 of thrombin, which is also the binding site for fibrinogen and (2) the amino terminal of bivalirudin binds to the active (catalytic) site of thrombin, responsible for catalytic conversion of fibrinogen to fibrin and the activation of clotting factors and platelets. After this initial high-affinity binding of bivalirudin to thrombin, there is an initial complete inhibition of both the fibrinogen-binding and catalytic functions of thrombin. However, thrombin molecule slowly starts an enzymatic cleavage of bivalirudin near the amino-terminal end resulting in release of aminoterminal segment from the catalytic site, leaving the catalytic site free to perform catalytic functions like activation of platelets, besides fibrin production (Figure 5). As a result of the partial cleavage of bivalirudin by thrombin, the binding of the carboxy-terminal fragment of bivalirudin molecule to exosite 1 of thrombin becomes low affinity and weakly competitive. Fibrinogen molecule is able to now displace this weakly bound bivalirudin fragment from the exosite 1 (fibrinogen-binding site) of thrombin and thereby able to present itself to the catalytic site of thrombin to be converted to fibrin. In summary, the initial binding of bivalirudin to thrombin causes a complete but temporary inhibition of thrombin. This is followed by slow cleavage of bivalirudin, leading to a slow recovery of thrombin function.5 The binding of bivalirudin to thrombin in a 1:1 ratio implies that each molecule of bivalirudin can only inhibit the action of a single molecule of thrombin (unlike heparin, which can free itself and become available for potentiating additional molecules of antithrombin). The gradual enzymatic cleavage of bivalirudin by thrombin thereby requires the need for a continuous infusion of bivalirudin to replenish the antithrombin activity of the drug to ensure therapeutic level of anticoagulation.6 In the segment of the vessel with stasis or slow flow, there is no availability of a fresh supply of bivalirudin molecules (that are unable to reach the local milieu distally) and thrombin molecules slowly recover from inhibition, leading to development of a prothrombotic milieu in the vasculature. Thrombin then initiates a thrombotic cascade that conceivably propagates proximally as thrombin amplifies its own production (overtaking the anticoagulant effect of the bivalirudin) leading to extensive coronary thrombosis. These situations are likely to occur during flow-limiting coronary www.americantherapeutics.com
e103
FIGURE 5. (A) Initial binding of bivalirudin to thrombin occurs at the exosite 1 and active (catalytic) site of thrombin molecule, causing complete inhibition of thrombin. (B) Thrombin slowly cleaves bivalirudin causing the release of the amino-terminal fragment of bivalirudin from the active site, allowing partial recovery of thrombin catalytic function (like platelet activation, besides fibrin production). (C) Finally, the fibrinogen molecule displaces the weakly bound bivalirudin from thrombin, allowing full recovery of thrombin function (including fibrin production).
dissection, placement of coronary hardware limiting coronary flow for a period of time (prolonged balloon inflations, difficult delivery of stents, bulky devices like brachytherapy catheters, etc.), and slow flow or no-reflow phenomenon. Experimental data support the hypothesis that the anticoagulant effect of bivalirudin is overcome rapidly American Journal of Therapeutics (2014) 21(4)
e104
in the situations of stasis, when there is no replenishment of the drug by infusion. An in vitro study compared the development of catheter thrombosis for bivalirudin, enoxaparin, and unfractionated heparin (UFH) in a controlled in vitro environment.7 Ten healthy male volunteers were pretreated with aspirin 500 mg 2 hours before venesection of 50 mL of blood. The 7 groups of anticoagulant combinations tested were as follows: UFH, UFH + eptifibatide, enoxaparin, enoxaparin + eptifibatide, bivalirudin bolus, bivalirudin + eptifibatide, and bivalirudin bolus + continuous infusion. The blood/anticoagulant mixture was continuously circulated through a cardiac guiding catheter for 60 minutes or until the catheter became blocked with thrombus. After anticoagulation with bolus dose bivalirudin, the catheter was invariably occluded with thrombus after 33 minutes of circulation. However, a continuous infusion of bivalirudin prevented the development of occlusive catheter thrombosis. In the bolus bivalirudin group, the mean thrombus weight was significantly greater than in all other groups (P , 0.01 in all analyses). Bivalirudin given as a bolus was not sufficient to prevent cardiac catheter thrombosis in this in vitro study. However, a continuous infusion of bivalirudin had similar antithrombotic efficacy compared with other treatment strategies. Similarly, there is a case report of a large thrombus forming in the venous reservoir while using bivalirudin during cardiopulmonary bypass in a patient with heparin-induced thrombocytopenia. This was presumably related to stasis of blood associated with the full venous reservoir maintained in the case leading to formation of a large thrombus at the top of the venous canister.8 In the REPLACE-2 trial,9 the investigators allowed addition of glycoprotein IIb/IIIa inhibitors (GPI) to patients randomized to bivalirudin, on a “provisional” basis, in the following circumstances where there was a likely perceived benefit with the use of the GPI: decreased thrombolysis in myocardial infarction flow (0–2) or slow reflow, dissection with decreased flow, new or suspected thrombus, persistent residual stenosis, distal embolization, unplanned stent, suboptimal stenting, side branch closure, abrupt closure, clinical instability, and prolonged ischemia. During the study, 1 or more of these circumstances occurred in 12.7% of the patients in the bivalirudin (with provisional GPI) arm. The GPIs were administered to 7.2% of the patients in the bivalirudin (with provisional GPIs) arm (62.2% of the eligible patients). The composite triple end point of death, MI, or urgent revascularization was higher (7.6%) in the bivalirudin with “provisional” GPI arm as compared with the heparin plus American Journal of Therapeutics (2014) 21(4)
Sharma et al
GPI arm (7.1%). If these “bailout” provisional utilization of GPIs had not been allowed, it is conceivable that the composite triple end point of death, MI, or urgent revascularization could have been much higher in the bivalirudin group, as studies have shown that intracoronary thrombosis in the setting of bivalirudin use is reversible with the use of intracoronary abciximab.4 Whether use of bivalirudin imparts a higher risk of acute thrombosis in certain settings needs more conclusive data, although in some trials, there are suggestions of an effect pointing to this direction. In the HORIZONS-AMI trial,2 acute stent thrombosis occurred more frequently in patients assigned to bivalirudin compared with heparin plus a GPI (1.4% vs. 0.3%, P , 0.001). In the ACUITY trial,1 comparing UFH plus GPIs, bivalirudin plus GPI, and bivalirudin alone, patients treated with bivalirudin alone had an increase in ischemic events, especially if they were troponin positive and were not pretreated with a thienopyridine. A review of the literature on the frequency of coronary thrombosis occurring in the setting of PCI with bivalirudin reveals case reports in the Web sites of drug safety and side effects. Many authors have reported anecdotal cases of the coronary thrombus formation during PCI with bivalirudin.4,10,11 Kuchulakanti et al report 2 cases of intracoronary thrombosis with gamma radiation when bivalirudin is used as an anticoagulant. The authors used an experimental swine model to replicate their clinical observations. There was occurrence of thrombus in the guide catheter and over the intervention guide wire in 2 of the swine in the bivalirudin group. They postulated that thrombus forms because of prolonged dwell time and stasis of blood in the catheter and may propagate into the coronary arteries in some cases.6 In the Brachytherapy and Bivalirudin Evaluation Study,12 of the 152 patients who underwent PCI and vascular brachytherapy with either gamma or beta radiation, there was a high prevalence of acute procedural intracoronary thrombosis in patients treated with gamma irradiation (25%). Although the incidence of this complication was much lower (0.7%, P , 0.001) in the beta-irradiation group, acute procedural intracoronary thrombosis did still occur with use of bivalirudin. Sharma et al reported occurrence of acute closure because of intracoronary thrombus formation during vascular brachytherapy with beta radiation with use of bivalirudin in 2 patients.4 Upon removal of the Novoste brachytherapy catheter after a dwell time of approximately 3 minutes, there was acute closure of the vessel distal to the treatment zone. The acute closure was reversed with use of intracoronary administration of abciximab, signifying the presence of fresh platelet www.americantherapeutics.com
Intracoronary Thrombosis With Bivalirudin
thrombus that was amenable to rapid dissolution with delivery of high local dose (with intracoronary administration) of abciximab. The reversible pharmacodynamics of bivalirudin demands an understanding on part of the interventionalist that a recovery of thrombin function can follow the catabolism and unbinding of bivalirudin from thrombin. This can occur rapidly when there is absence of a replenishing delivery of the drug because of stasis or slow flow. This can tip the local milieu from an antithrombotic state toward a prothrombotic state locally at the intervention site and lead to development of extensive intracoronary thrombus formation with a high risk of mortality, as occurred in these 2 cases. We believe that prolonged PCI procedures associated with a situation of decreased TIMI flow (0–2) or slow reflow, flow-limiting dissection, new or suspected thrombus, distal embolization, and prolonged dwell times with any intervention hardware, especially bulky devices like brachytherapy catheter, can predispose to formation of profuse coronary thrombus with use of bivalirudin during PCI, with a risk of death. It demands an immediate treatment strategy with possibly a switch to heparin and a judicious use of GPIs in these situations.
CONCLUSIONS Extensive intracoronary thrombus formation with a high risk of mortality can follow prolonged PCI procedures associated with a situation of stasis and decreased TIMI flow (0–2) or slow reflow, flow-limiting dissection, and prolonged dwell times with any intervention hardware with use of bivalirudin because of its reversible kinetics of inhibition of thrombin. It demands an immediate recognition by the interventionalist, with immediate treatment strategy with possibly a switch to heparin and a judicious use of GPIs.
ACKNOWLEDGMENTS The authors would like to express their thanks to Mark Monterroso and Rohan Sharma for their assistance in preparation of the manuscript and figures. The authors also thank Dr Rasham Sandhu and Dr Brijesh Bhambi for their critical input in formulation of the final manuscript.
www.americantherapeutics.com
e105
REFERENCES 1. Stone GW, White HD, Ohman EM, et al; Acute Catheterization and Urgent Intervention Triage strategy (ACUITY) trial investigators. Bivalirudin in patients with acute coronary syndromes undergoing percutaneous coronary intervention: a subgroup analysis from the Acute Catheterization and Urgent Intervention Triage strategy (ACUITY) trial. Lancet. 2007;369:907–919. 2. Stone GW, Witzenbichler B, Guagliumi G, et al; HORIZONS-AMI Trial Investigators. Bivalirudin during primary PCI in acute myocardial infarction. N Engl J Med. 2008;358:2218–2230. 3. Angiomax (bivalirudin). Package Insert. Parsippany, NJ: The Medicines Company; 2010. 4. Sharma S, Bhambi B, Nyitray W, et al. Bivalirudin (Angiomax) use during intracoronary brachytherapy may predispose to acute closure. J Cardiovasc Pharmacol Ther. 2003;8:9–15. 5. Sciulli TM, Mauro VF. Pharmacology and clinical use of bivalirudin. Ann Pharmacother. 2002;36:1028–1041. 6. Kuchulakanti PK, Satler LF, Rha SW, et al. Bivalirudinassociated intracoronary thrombosis during gammabrachytherapy and its experimental validation in acute swine model. Catheter Cardiovasc Interv. 2004;62:209–213. 7. Maegdefessel L, Buerke M, Schubert S, et al. Comparison of bivalirudin, enoxaparin, and unfractionated heparin in preventing cardiac catheter thrombosis. Results of an in-vitro study. Thromb Haemost. 2008;100:693–698. 8. Wong JK, Tian Y, Shuttleworth P, et al. Case report: a thrombus in the venous reservoir while using bivalirudin in a patient with heparin-induced thrombocytopenia undergoing heart transplantation. Anesth Analg. 2010; 111:609–612. 9. Lincoff AM, Bittl JA, Harrington RA, et al; REPLACE-2 Investigators. Bivalirudin and provisional glycoprotein IIb/IIIa blockade compared with heparin and planned glycoprotein IIb/IIIa blockade during percutaneous coronary intervention: REPLACE-2 randomized trial. JAMA. 2003;289:853–863. 10. Tadros GM, Broder K, Bachour F. Intracoronary macrothrombus formation during percutaneous coronary intervention despite optimal activated clotting time using bivalirudin—a case report. Angiology. 2005;56: 761–765. 11. Chieffo A, Melzi G, Rogacka R, et al. Safety and feasibility of Bivalirudin with either Cypher and Taxus drugeluting stent during percutaneous coronary intervention. EuroIntervention. 2005;1:70–74. 12. Kuchulakanti P, Wolfram R, Torguson R, et al. Brachytherapy and bivalirudin evaluation study. Am Heart J. 2005;150:832–837.
American Journal of Therapeutics (2014) 21(4)
