Bolus Thrombolysis in Venous Thromboembolism* Giancarlo AgneUi, M.D., and Pasquale Parise, M.D.
Thrombolytic therapy is rarely used in venous thromboembolism because of the fear ofhemorrhagic complications. Preliminary clinical experiences with recombiDaDt tissuetype plasminogen activator (rt-PA) in patients with deep vein thrombosis have shown that even this Obrin-specific plasminogen activator causes an unacceptable rate of hemorrhagic complications. Theoretical considerations and the available experimental and clinical data suggest that infusion of rt-PA over a short period of time would result in a more favorable risk-benefit ratio. Shortening the period of rt-PA infusion results in higher peak plasma levels, thus allowing a higher concentration of the plasminogen activator on the surface and inside the occluding thrombus. In addition, a bolus infusion can prevent or minimize the interaction between rt-PA and the hemostatic system, reducing the likelihood of a systemic lytic state, of a platelet function defect, and, possibly, of bleeding side effects. In venous thromboembolism animal models, the efficacy of bolus rt-PA can be further increased by the adjunctive administration of an effective antithrombotic treabnent. This is because the accretion of new Obrin on the thrombi counteracts the lysis of preformed Obrin and in8uences negatively the Goal thrombus size. Effective adjunctive antithrombotic treatment includes either high doses of heparin, producing an unclottable activated partial thromboplastin time (aYfT), or doses of recombiDaDt hirudin, doubling the aYfT. When used as an alternative to rt-PA, bolus doses of a hybrid plasminogen activator with prolonged half-life efficiently reduce thrombus size by lysing preformed and newly formed 6brin. Preliminary clinical experience in patients with pulmonary embolism seems to con6rm that rt-PA infused as a bolus is at least as effective as, and probably more effective than, rt-PA infused over a longer period.
In recent years, thrombolytic therapy has become widely used in patients with acute myocardial infarction. Indeed, thrombolytic agents have been shown to induce reperfusion of the occluded coronary arteries, save ischemic myocardium, improve left ventricular dysfunction, and enhance both early and late survival.':" In contrast, the use of thrombolytic agents remains controversial and infrequent in other thromboembolic diseases, such as venous thromboembolism, peripheral obstructive arterial disease, and unstable angina, mainly because of the fear of a negative benefit-risk ratio." *From the Istituto di Medicina Intema e Medicina Vascolare, UniversitA di Perugia, Perugia, Italy. Reprint requests: Dr AgneUi, Istituto lntema, Via Enrico dallbzzo, ImJgia, Italy 06100
1728
Most experience with thrombolytic agents in venous thromboembolism is with streptokinase and urokinase, the first-generation plasminogen activators. Streptokinase and urokinase have little affinity for fibrin. They activate both circulating and thrombusbound plasminogen, which results in systemic plasmin generation and, consequently, in a severe hemostatic defect, which could contribute to the hemorrhagic complications of thrombolytic therapy:" TIssue-type plasminogen activator (t-PA), a second-generation thrombolytic agent, has high affinity for fibrin and is activated by fibrin. In the absence of fibrin, tepA activates plasminogen at a very slow rate; however, in the presence of fibrin or fibrin fragments, the rate of plasminogen activation is accelerated a hundredfold." Studies in animal models have shown that recombinant tepA (rt-PA), in contrast with streptokinase, has the potential to induce thrombolysis without producing a systemic lytic state. 6 •7 Thus, it was originally hoped that the avidity oft-PA for fibrin and its modest impact on the hemostatic system would result in fewer bleeding complications than occur with first-generation thrombolytic agents. However, both animal and clinical studies have demonstrated that the thrombus specificity of rt-PA is dose-dependent and that rt-PA doses required for effective and rapid thrombolysis induce systemic fibrinolysis. 7- 1O In most clinical studies, rt-PA was administered by continuous infusion over a period of 90 min to 8 h. The rationale for such a prolonged rt-PA infusion was based on the fear of a high incidence of rethrombosis due to (1) the moderate hemostatic impairment induced by rt-PA; (2) the short half-life of t-PA; and (3) the general conviction that the affinity for fibrin and the ability oft-PA to be activated by fibrin would be sufficient to provide selective thrombolysis. However, currently adopted regimens of rt-PA administration induce a plasma fibrinolytic state, 11 and although they produce less fibrinogenolysis than streptokinase for an equivalent thrombolytic effect,12 they are associated with excessive bleeding. Furthermore, there is no clear evidence that rt-PA, with the adopted regimens, has a higher clinical efficacy than streptokinase or urokinase. 1.2.13 Experimental observations and clinical data suggest that different strategies in administering rt-PA could increase both efficacy and safety. Among these strategies is the bolus administration of rt- PA. Biochemical and biologic properties of rt-PAlet us foresee potential BolusThrombolysis in venous Thromboembolism (Agnelli, Parise)
advantages of this agent, when given as a bolus, over other plasminogen activators. For these reasons, most of this review; dealing with bolus administration of thrombolytic agents, will be dedicated to this agent. RATIONALE FOR BOLUS RT-PA ADMINISTRATION
Theoretically, one way to improve thrombolysis is to reach a higher concentration of the lytic agent in the vicinity of the thrombus. The efficacy of intraarterially administered thrombolytic agents in coronary or peripheral artery obstruction is presumably due to the high concentration of plasminogen activators on the surface and inside the occluding thrombus. 2 Since it is not possible to increase the total dose of administered rt-PA because of an unacceptable risk of bleeding, 14 the only possible way to obtain higher peak plasma levels of rt-PA is to shorten the infusion time. The effectiveness and the safety of such an approach are supported by a number of experimental and clinical observations. In a rabbit jugular vein model with radioactive thrombi;" the infusion of 0.6 mg ofrt-PA per kilogram of body weight over 15, 30, or 60 min resulted in a significantly higher thrombolytic effect than the infusion of the same dose over 4 h. In addition, the 30and 15-min rt-PA infusions, in contrast with 1- and 4-h infusions, did not produce significantly more bleeding than a saline infusion and did not cause any reduction in plasma levels of ~-antiplasmin. These findings indicate that a rapid-infusion regimen results in improved thrombolysis with minimal fibrinogenolysis and without excessive bleeding. Shiffman et al16 investigated the effects of 2 dosing regimens of rt-PA on pulmonary thrombolysis in a canine model of pulmonary embolism that caused pulmonary hypertension. In that study, rt-PA, 1 mgl kg, was infused over 15 or 90 min. The shorter infusion resulted in a marked increase in the rate of lysis (56%/h vs 27%/h, p
Thrombolytic therapy is rarely used in venous thromboembolism because of the fear ofhemorrhagic complications. Preliminary clinical experiences with recombiDaDt tissuetype plasminogen activator (rt-PA) in patients with deep vein thrombosis have shown that even this Obrin-specific plasminogen activator causes an unacceptable rate of hemorrhagic complications. Theoretical considerations and the available experimental and clinical data suggest that infusion of rt-PA over a short period of time would result in a more favorable risk-benefit ratio. Shortening the period of rt-PA infusion results in higher peak plasma levels, thus allowing a higher concentration of the plasminogen activator on the surface and inside the occluding thrombus. In addition, a bolus infusion can prevent or minimize the interaction between rt-PA and the hemostatic system, reducing the likelihood of a systemic lytic state, of a platelet function defect, and, possibly, of bleeding side effects. In venous thromboembolism animal models, the efficacy of bolus rt-PA can be further increased by the adjunctive administration of an effective antithrombotic treabnent. This is because the accretion of new Obrin on the thrombi counteracts the lysis of preformed Obrin and in8uences negatively the Goal thrombus size. Effective adjunctive antithrombotic treatment includes either high doses of heparin, producing an unclottable activated partial thromboplastin time (aYfT), or doses of recombiDaDt hirudin, doubling the aYfT. When used as an alternative to rt-PA, bolus doses of a hybrid plasminogen activator with prolonged half-life efficiently reduce thrombus size by lysing preformed and newly formed 6brin. Preliminary clinical experience in patients with pulmonary embolism seems to con6rm that rt-PA infused as a bolus is at least as effective as, and probably more effective than, rt-PA infused over a longer period.
In recent years, thrombolytic therapy has become widely used in patients with acute myocardial infarction. Indeed, thrombolytic agents have been shown to induce reperfusion of the occluded coronary arteries, save ischemic myocardium, improve left ventricular dysfunction, and enhance both early and late survival.':" In contrast, the use of thrombolytic agents remains controversial and infrequent in other thromboembolic diseases, such as venous thromboembolism, peripheral obstructive arterial disease, and unstable angina, mainly because of the fear of a negative benefit-risk ratio." *From the Istituto di Medicina Intema e Medicina Vascolare, UniversitA di Perugia, Perugia, Italy. Reprint requests: Dr AgneUi, Istituto lntema, Via Enrico dallbzzo, ImJgia, Italy 06100
1728
Most experience with thrombolytic agents in venous thromboembolism is with streptokinase and urokinase, the first-generation plasminogen activators. Streptokinase and urokinase have little affinity for fibrin. They activate both circulating and thrombusbound plasminogen, which results in systemic plasmin generation and, consequently, in a severe hemostatic defect, which could contribute to the hemorrhagic complications of thrombolytic therapy:" TIssue-type plasminogen activator (t-PA), a second-generation thrombolytic agent, has high affinity for fibrin and is activated by fibrin. In the absence of fibrin, tepA activates plasminogen at a very slow rate; however, in the presence of fibrin or fibrin fragments, the rate of plasminogen activation is accelerated a hundredfold." Studies in animal models have shown that recombinant tepA (rt-PA), in contrast with streptokinase, has the potential to induce thrombolysis without producing a systemic lytic state. 6 •7 Thus, it was originally hoped that the avidity oft-PA for fibrin and its modest impact on the hemostatic system would result in fewer bleeding complications than occur with first-generation thrombolytic agents. However, both animal and clinical studies have demonstrated that the thrombus specificity of rt-PA is dose-dependent and that rt-PA doses required for effective and rapid thrombolysis induce systemic fibrinolysis. 7- 1O In most clinical studies, rt-PA was administered by continuous infusion over a period of 90 min to 8 h. The rationale for such a prolonged rt-PA infusion was based on the fear of a high incidence of rethrombosis due to (1) the moderate hemostatic impairment induced by rt-PA; (2) the short half-life of t-PA; and (3) the general conviction that the affinity for fibrin and the ability oft-PA to be activated by fibrin would be sufficient to provide selective thrombolysis. However, currently adopted regimens of rt-PA administration induce a plasma fibrinolytic state, 11 and although they produce less fibrinogenolysis than streptokinase for an equivalent thrombolytic effect,12 they are associated with excessive bleeding. Furthermore, there is no clear evidence that rt-PA, with the adopted regimens, has a higher clinical efficacy than streptokinase or urokinase. 1.2.13 Experimental observations and clinical data suggest that different strategies in administering rt-PA could increase both efficacy and safety. Among these strategies is the bolus administration of rt- PA. Biochemical and biologic properties of rt-PAlet us foresee potential BolusThrombolysis in venous Thromboembolism (Agnelli, Parise)
advantages of this agent, when given as a bolus, over other plasminogen activators. For these reasons, most of this review; dealing with bolus administration of thrombolytic agents, will be dedicated to this agent. RATIONALE FOR BOLUS RT-PA ADMINISTRATION
Theoretically, one way to improve thrombolysis is to reach a higher concentration of the lytic agent in the vicinity of the thrombus. The efficacy of intraarterially administered thrombolytic agents in coronary or peripheral artery obstruction is presumably due to the high concentration of plasminogen activators on the surface and inside the occluding thrombus. 2 Since it is not possible to increase the total dose of administered rt-PA because of an unacceptable risk of bleeding, 14 the only possible way to obtain higher peak plasma levels of rt-PA is to shorten the infusion time. The effectiveness and the safety of such an approach are supported by a number of experimental and clinical observations. In a rabbit jugular vein model with radioactive thrombi;" the infusion of 0.6 mg ofrt-PA per kilogram of body weight over 15, 30, or 60 min resulted in a significantly higher thrombolytic effect than the infusion of the same dose over 4 h. In addition, the 30and 15-min rt-PA infusions, in contrast with 1- and 4-h infusions, did not produce significantly more bleeding than a saline infusion and did not cause any reduction in plasma levels of ~-antiplasmin. These findings indicate that a rapid-infusion regimen results in improved thrombolysis with minimal fibrinogenolysis and without excessive bleeding. Shiffman et al16 investigated the effects of 2 dosing regimens of rt-PA on pulmonary thrombolysis in a canine model of pulmonary embolism that caused pulmonary hypertension. In that study, rt-PA, 1 mgl kg, was infused over 15 or 90 min. The shorter infusion resulted in a marked increase in the rate of lysis (56%/h vs 27%/h, p
