Current Status of Thrombolysis for Peripheral Arterial Occlusive Disease

Developments in Endovascular and Endoscopic Surgery SECTION EDITOR: Samuel S. Ahn, MD

Current Status of Thrombolysis for Peripheral Arterial Occlusive Disease Kenneth Ouriel, MD, Cleveland, Ohio

Acute peripheral arterial occlusion occurs as a result of thrombosis or embolism. A reduction in the prevalence of rheumatic heart disease accounts for a shift in the frequency of embolic to thrombotic occlusions. Also, a dramatic increase in the number of lower extremity arterial bypass graft procedures explains the predominance of graft occlusions in most recent series of patients with acute limb ischemia. While open surgical procedures remain the gold standard in the treatment of peripheral arterial occlusion, thrombolytic agents have been employed as an alternative to primary surgical revascularization in patients with acute limb ischemia. Systemic administration of thrombolytic agents, while effective for small coronary artery clots, fails to achieve dissolution of the large peripheral arterial thrombi. Catheter-directed administration of the agents directly into the occlusive thrombus is the only means of effecting early recanalization. Prior to 1999, urokinase was the sole agent used in North America for peripheral arterial indications, but the loss of the agent from the marketplace forced clinicians to turn to alternate agents, specifically alteplase and reteplase. Interest in the use of platelet glycoprotein inhibitors and mechanical thrombectomy devices also rose, coincident with the loss of urokinase from the marketplace. Most clinicians welcome the predicted return of urokinase to the marketplace. New investigative trials should be organized and executed to answer some of the remaining questions related to thrombolytic treatment of peripheral arterial disease. Foremost in this regard remains the question of which patients are best treated with percutaneous thrombolytic techniques and which are best treated with primary operative intervention. Ultimately, however, the thrombolytic agents are but one tool in the armamentarium of the vascular practitioner. This review is directed at providing the practicing clinician with the basic fund of knowledge necessary when determining the most appropriate intervention in a particular patient with peripheral arterial occlusion, be it thrombolytic therapy, percutaneous mechanical thrombectomy, primary surgical revascularization, or a combination of the three.

INTRODUCTION Acute occlusion of a native artery or bypass graft is associated with the immediate development of pain in the extremity. When severe, pain is folDepartment of Vascular Surgery, The Cleveland Clinic Foundation, Cleveland, OH. Correspondence to: K. Ouriel, MD, Department of Vascular Surgery, Desk S40, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, OH 44195, USA, E-mail: [email protected]. Ann Vasc Surg 2002; 16: 797-804 DOI: 10.1007/s10016-001-0318-y Ó Annals of Vascular Surgery Inc. Published online: 24 October 2002

lowed by paresthesia, motor dysfunction, and, eventually, tissue infarction. The rapidity of onset is correlated with the extent of the thrombotic process and the amount of preexisting collateral pathways. Whether from in situ thrombosis of a native artery or bypass graft or from embolization, the acute limb ischemia threatens both the patient’s limb and life. Mortality is closely linked with the presence of medical comorbidities such as coronary artery disease rather than directly associated with the ischemic process itself, and in-hospital mortality rates exceed 20% in this group of individuals.1,2 797

798 Ouriel

Treatment begins with early heparin anticoagulation to limit the propagation of thrombus and prevent clinical deterioration, although there are few objective data on which to base this practice.1,3,4 Traditionally, anticoagulation has been followed by urgent surgical intervention, employing thromboembolectomy, placement of a bypass graft, or other techniques to restore arterial flow to the extremity. Early operative intervention, however, is associated with a significant risk of perioperative mortality. The classic study of Blaisdell et al., published in the late 1970s, documented mortality rates in excess of 25% following open surgical repair for acute leg ischemia.1 Later, Jivega˚rd and colleagues corroborated these findings, documenting a 20% mortality rate in patients undergoing operative revascularization for acute limb ischemia.2 Despite advances in surgical technique and perioperative care, the risk of morbidity and mortality following open surgical intervention continues to be significant.5-8 There are several factors that explain this observation, but the compromised medical status of the patients who present with acute peripheral arterial occlusion is foremost. Patients who develop acute limb ischemia are often elderly, with advanced systemic atherosclerosis. They are poorly equipped to tolerate the insult of ischemia of an extremity, and early operation without the benefit of medical stabilization accounts for the high rate of complications. The literature confirms that individuals who present with acute, limb-threatening ischemia comprise one of the sickest subgroups of patients that the peripheral vascular practitioner is asked to treat,9 patients for whom early intervention is necessary to salvage an extremity but who are ill equipped to tolerate an invasive treatment modality. Thrombolytic therapy offers a less invasive option in this group of patients.2 Intraarterial thrombolytic therapy has gained prominence as an initial intervention, in which thrombolytic agents are infused directly into the occluding thrombus through a catheter-directed approach. Agents such as urokinase,10 alteplase,11 and reteplase12 can restore adequate arterial perfusion and subsequent arteriographic studies allow the clinician to identify and address the culprit lesions responsive for the occlusion. Often an endovascular procedure can be performed to minimize the risk to the patients. In other cases where open surgical intervention is still necessary, it can be performed on an elective basis in a well-prepared patient.13 While theoretically attractive, thrombolytic therapy has been criticized, as a high rate of reocclusion, prohibitive cost, and inferior long-term

Annals of Vascular Surgery

patency have been cited.14,15 While based on carefully documented observation, some of the criticisms have implied an improper understanding of therapeutic expectations. The need for subsequent intervention to address unmasked lesions was often neglected. Experience has demonstrated that thrombolysis must be followed by definitive therapy to address the underlying lesion that caused the occlusion. In fact, when no such lesion can be found, the risk of early rethrombosis is unacceptably high.16 As testimony to this caveat, Sullivan observed post-thrombolytic 2-year patency rates of 79% in bypass grafts with flow-limiting lesions identified and corrected by angioplasty or surgery, in contrast to only 9.8% in those without such lesions.

THE THROMBOLYTIC AGENTS All clinically available thrombolytic agents in clinical use are plasminogen activators (Table I). As such, they do not directly degrade fibrinogen. Rather, they are trypsin-like serine proteases that have high, specific activity directed at the cleavage of a single peptide bond in the plasminogen zymogen, converting it to plasmin. Plasmin is the active molecule that cleaves fibrin polymer to cause the dissolution of thrombus. Milstone first recognized the importance of plasminogen in 1941, when it was noted that clots formed with highly purified fibrinogen and thrombin were not lysed by streptococcal fibrinolysin unless a small amount of human serum (plasminogen) was added.17 Recognizing this direct role of plasminogen, early investigators attempted to dissolve occluding thrombi with the administration of exogenous plasmin. Free plasmin, however, was ineffective as a thrombolytic agent. Plasmin is extremely unstable at physiologic pH, with autodegradation accounting for the failure of these attempts. Effective thrombolysis can only be achieved when fibrinbound plasminogen is converted to its active form plasmin at the site of the thrombus. The dependence of fibrinolysis on adequate circulating levels of plasminogen is best illustrated by studies of the fibrinolytic potential of blood drawn from patients receiving intravenous administration of thrombolytic agents for acute myocardial infarction.18 Blood obtained soon after the start of thrombolytic administration displayed a great degree of in vitro fibrinolytic potential. Aliquots of plasma drawn from the patients and then added to radiolabeled clots in test tubes produced rapid dissolution of the clots. By contrast, similar aliquots

Vol. 16, No. 6, 2002

Thrombolysis for peripheral arterial occlusive 799

Table I. Properties of thrombolytic agents Component

Molecular Weight

Plasma t1/2 (min)

Properties

Streptokinase Urokinase r-urokinase Prourokinase rt-PA TNK-tPA Reteplase

48,000 32,000/54,000 32,000 49,000 68,000 65,000 39,000

16/90 14 7 7 3.5 15 14

Complexes with plasminogen to gain activity Direct plasminogen activator Similar in most respects to natural urokinase Little intrinsic activity, converted to urokinase Greater fibrin anity and specificity A modified t-PA with a longer half-life A truncated t-PA with a longer half-life

drawn from patients after 20 min thrombolytic administration had considerably less thrombolytic potential. The explanation for this observation relates to the amount of plasminogen present in the blood. Prolonged thrombolytic administration consumed all of the endogenous plasminogen and, despite continued administration of thrombolytic agent, no further clot lysis was possible.

These early studies were followed by reports on the use of SK in patients with occluding vascular thrombi. In 1956, E.E. Cliffton, at the Cornell University Medical College in New York, was responsible for the first brief description of the clinical effectiveness of intravascular thrombolytic administration.22 The following year, Cliffton published his results in 40 patients with occlusive thrombi treated with a SK-plasminogen in combination.22 The location of the thrombi was diverse, and included peripheral arterial thrombi, venous thrombi, pulmonary emboli, retinal occlusions, and, in two patients, occlusive carotid thrombi. Cliffton’s clinical results were far from exemplary, recanalization was not uniform, and bleeding complications were frequent. Nevertheless, he must be credited with the first use of thrombolytic agents for the treatment of pathologic thrombi, as well as with the first use of catheter-directed administration of a thrombolytic agent.

History of Thrombolytic Therapy In 1933, Tillett and Garner at the Johns Hopkins Medical School discovered that filtrates of broth cultures of certain strains of hemolytic streptococcus bacteria had fibrinolytic properties.19 This streptococcal byproduct was originally termed streptococcal fibrinolysin. The purity of this agent was poor, however. Clinical use, of necessity, awaited adequate purification. Tillett and Sherry administered streptokinase (SK) intrapleurally to dissolve loculated hemothoraces in the late 1940s, but intravascular administration was not attempted until the following decade.20 Tillett first reported intravascular administration of a thrombolytic agent in an article published in 1955.21 A concentrated and partially purified SK (VaridaseÒ, Lederle Laboratories) was injected into 11 patients. This investigation was performed with the intent to gain data on the safety of the agent in volunteers; in no case was the SK administered to dissolve pathologic thrombi. Fever and hypotension developed as the amount of SK approached therapeutic levels. Whereas fever was generally mild and controllable with antipyretics, hypotension was sometimes prominent. The mean fall in systolic pressure was 31 mmHg and three of the patients manifested systolic pressures below 80 mmHg. These untoward reactions were more likely a result of contaminants in the preparation rather than the SK itself. Despite these reactions, systemic proteolysis was observed, with a decrease in fibrinogen and plasminogen, concurrent with a mild increase in the prothrombin time.

CLASSIFICATION OF THROMBOLYTIC AGENTS Several schemes may be used to classify thrombolytic agents. The agents may be grouped by their mechanism of action—those that directly convert plasminogen to plasmin versus those that are inactive zymogens and require transformation to an active form before they can cleave plasminogen. Thrombolytic agents can be grouped by their mode of production—those that are manufactured via recombinant techniques and those that are of bacterial origin. Of interest, recombinant agents harvested from a bacterial expression system such as Escherichia coli do not contain carbohydrates, while products of mammalian hybridoma (e.g., recombinant prourokinase from mouse hybridoma SP2/0 cells) are fully glycosylated. Thrombolytic agents can be classified by their pharmacologic actions—those that are ‘‘fibrin-specific’’ (bind to fi-

800 Ouriel

brin but not fibrinogen) as opposed to nonspecific, and those that have a great degree of ‘‘fibrin-affinity’’ (bind avidly to fibrin) as opposed to those that do not. We have found it most useful to classify thrombolytic agents into groups based on the origin of the parent compound. It is most efficient to divide the agents into four groups: the streptokinase compounds, the urokinase compounds, the tissue plasminogen activators, and an additional, miscellaneous group consisting of novel agents that are distinct from agents in the three other groups. Streptokinase Compounds Streptokinase, originating from the streptococcus bacteria, was the first thrombolytic agent to be described.19 SK has a biphasic half-life comprising a rapid t1/2 of 16 min and a second, slower t1/2 of 90 min. Whereas the initial half-life is accounted for by the formation of a complex between the molecule and SK antibodies, the second half-life represents the actual biologic elimination of the protein. SK differs from other thrombolytic agents with respect to the stoichiometry of plasminogen binding. Whereas other agents directly convert plasminogen to plasmin, SK must form an equimolar stoichiometric complex with a plasmin or plasminogen molecule to gain activity. Only then can this SKplasmin(ogen) complex activate a second plasminogen molecule to form active plasmin; thus, two plasminogen molecules are utilized in SK-mediated plasmin generation. SK suffers from the limitation of antigenic potential. Preformed antibodies exist to a certain extent in all patients who have been infected with the Streptococcus bacterium. Similarly, patients with exposure to SK may have high antibody titers on repeat exposure. These neutralizing antibodies inactivate exogenously administered streptokinase. SK antibodies may be overwhelmed through the use of a large initial bolus of drug, and a large initial SK loading dose may be employed in this regard. Some investigators have recommended measurement of antibody titers prior to beginning SK therapy, gauging the loading dose on the basis of this titer.23 While SK administration may be complicated by allergic reactions such as urticaria, pyrexia, periorbital edema, and bronchospasm, the major untoward effect associated with SK is hemorrhage. SK-associated hemorrhage may be no different from bleeding associated with any thrombolytic agent. The primary cause is likely the actions of systemic agent on the thrombi sealing the sites of vascular defects. The generation of free plasmin, however, can contribute to the problem,

Annals of Vascular Surgery

with degradation of fibrinogen and other serum clotting proteins, as well as the release of fibrin(ogen)-degradation products that are potent anticoagulants that themselves exacerbate the coagulopathy. Urokinase Compounds The fibrinolytic potential of human urine was first described by Macfarlane and Pinot in 1947.24 The active molecule was extracted, isolated, and named ‘‘urokinase’’ (UK) in 1952.25 The high-molecularweight form predominates in UK isolated from urine, while the low-molecular-weight form is found in UK obtained from tissue culture of kidney cells. Unlike SK, UK directly activates plasminogen to form plasmin; prior binding to plasminogen or plasmin is not necessary for activity. Also in contrast to SK, preformed antibodies to UK are not observed. The agent is nonantigenic and untoward reactions of fever or hypotension are rare. The most commonly employed UK in the United States has been of tissue-culture origin, manufactured from human neonatal kidney cells (AbbokinaseÒ, Abbott Laboratories, Abbott park, IL). UK has been fully sequenced, and a recombinant form of UK (r-UK) was tested in a single trial of patients with acute myocardial infarction and in two multicenter trials of patients with peripheral arterial occlusion.26 r-UK is fully glycosylated, since it is derived from a murine hybridoma cell line. r-UK differs from Abbokinase in several respects. First, rUK has a higher molecular weight than Abbokinase. Second, r-UK has a shorter half-life than its low-molecular-weight counterpart. Despite these differences, however, the clinical effects of the two agents have been quite similar. A precursor of UK was discovered in urine in 1979.27 Prourokinase was characterized and subsequently manufactured by recombinant technology using Escherichia coli (nonglycosylated) or mammalian cells (fully glycosylated). This singlechain form is an inactive zymogen, inert in plasma, but can be activated by kallikrein or plasmin to form active two-chain UK. This property accounts for amplification of the fibrinolytic process—as plasmin is generated, more prourokinase is converted to active urokinase, and the process is repeated. Prourokinase is relatively fibrin-specific, that is, its fibrin-degrading (fibrinolytic) activity greatly outweighs its fibrinogen-degrading (fibrinogenolytic) activity. This feature is explained by the preferential activation of fibrin-bound plasminogen found in a thrombus over free plasminogen found in flowing blood. Nonselective activators such as

Vol. 16, No. 6, 2002

SK and UK activate free and bound plasminogen equally and induce systemic plasminemia with resultant fibrinogenolysis and degradation of factors V and VII. Given the potential advantages of prourokinase (ProUK) over UK, Abbott Laboratories produced a recombinant form of proUK (r-ProUK) from a murine hybridoma cell line. This recombinant agent, named ProlyseÒ (Abbott Laboratories), is converted to active two-chain UK by plasmin and kallikrein. Prolyse and has been studied in the settings of myocardial infarction, stroke, and peripheral arterial occlusion. To date, it appears that r-ProUK offers the advantages associated with an agent that does not originate from a human cell source.8 Tissue Plasminogen Activators Tissue plasminogen activator (t-PA) is a naturally occurring fibrinolytic agent produced by endothelial cells and intimately involved in the balance between intravascular thrombogenesis and thrombolysis.28 Natural t-PA is a single-chain (527 amino acid) serine protease. In contrast to most serine proteases (e.g., UK), the single-chain form of t-PA has significant activity. t-PA has potential benefits over other thrombolytic agents. The agent exhibits significant fibrin specificity. In plasma, the agent is associated with little plasminogen activation. At the site of the thrombus, however, the binding of t-PA and plasminogen to the fibrin surface induces a conformational change in both molecules, greatly facilitating the conversion of plasminogen to plasmin and dissolution of the clot. t-PA also manifests the property of fibrin affinity, that is, it binds strongly to fibrin. Other fibrinolytic agents such as prourokinase do not share this property of fibrin affinity. Recombinant t-PA (rt-PA) was produced in the 1980s after molecular cloning techniques were used to express human t-PA DNA. ActivaseÒ (Genentech), a predominantly single-chain form of rt-PA, was eventually approved in the United States for the indications of acute myocardial infarction and massive pulmonary embolism. rt-PA has been studied extensively in the setting of coronary occlusion. In the GUSTO-I study of approximately 41,000 patients with acute myocardial infarction, rt-PA was more effective than SK in achieving vascular patency.29 Despite a slightly greater risk of intracranial hemorrhage with rt-PA, overall mortality was significantly reduced. In an effort to lengthen the duration of bioavailability of t-PA, the molecule was systemati-

Thrombolysis for peripheral arterial occlusive 801

cally bioengineered. Initial investigations identified regions in kringle 1 and the protease portion of tPA that mediate hepatic clearance, fibrin specificity, and resistance to plasminogen activator inhibitor. Three sites were modified to create TNK-tPA, a novel molecule with a greater half-life and fibrin specificity. The longer half-life of TNK-tPA allowed successful administration as a single bolus, in contrast to the requirement for an infusion with rt-PA. In addition, TNK-tPA manifests greater fibrin specificity than rt-PA, resulting in less fibrinogen depletion. In studies of acute coronary occlusion, TNK-tPA performed at least as well as rt-PA, concurrent with greater ease of administration.30 Similar to TNK-tPA, the novel recombinant plasminogen activator reteplase comprises the kringle 2 and protease domains of t-PA. Reteplase was developed with the goal of avoiding the necessity of a continuous intravenous infusion, thereby simplifying ease of administration. Reteplase (RetavaseÒ, Centocor), produced in Escherichia coli cells, is nonglycosylated, demonstrating a lower fibrin-binding activity and a diminished affinity to hepatocytes. This latter property accounts for a longer half-life than that of rt-PA, potentially enabling bolus injection instead of prolonged infusion. The fibrin affinity of reteplase was only 30% that exhibited with t-PA, similar to UK. The decrease in fibrin affinity was hypothesized to reduce the incidence of distant bleeding complications, in a manner similar to that of SK over rt-PA in the GUSTO trial.30 In fact, several properties of reteplase may account for a decreased risk of hemorrhage, including poor lysis of platelet-rich, older clots.31 Reteplase has been studied in several small trials, and its safety and efficacy appear to be similar to that of alteplase.12,32

THROMBOLYTIC CLINICAL TRIALS There have been three well-controlled, randomized comparisons of thrombolytic therapy to primary operation in patients with recent peripheral arterial occlusion. The first study, the Rochester series, compared UK therapy to primary operation in a single-center experience of 114 patients presenting with what has subsequently been called ‘‘hyperacute ischemia.’’ 5 Enrolled patients in this trial all had severely threatened limbs (Rutherford class IIb) with a mean symptom duration of approximately 2 days. After 12 months of follow-up, 84% of patients randomized to UK were alive, compared to only 58% of patients randomized to primary operation (Fig. 1). By contrast, the rate of limb

802 Ouriel

Annals of Vascular Surgery

Fig. 1. Rochester trial 12-month follow-up data.

salvage was identical at 80%. A closer inspection of the raw data revealed that the defining variable for mortality differences was the development of cardiopulmonary complications during the periprocedural period. The rate of long-term mortality was high when such periprocedural complications occurred but was relatively low when they did not occur. It was only the fact that such complications occurred more commonly in patients taken directly to the operating theater that explained the greater long-term mortality rate in the operative group. The second prospective, randomized analysis of thrombolysis versus surgery was the Surgery or Thrombolysis for the Ischemic Lower Extremity (STILE) trial. Genentech (South San Francisco, CA), the manufacturer of the Activase brand of rtPA, funded the study. At its termination, 393 patients were randomized to one of three treatment groups: rt-PA, UK, or primary operation. Subsequently, the two thrombolytic groups were combined for purposes of data analysis when the outcome was found to be similar. While the rate of the composite end point of untoward events was

Fig. 2. Outcome measures from the STILE data after 30 days of follow-up. Note that the rates of death and amputation are similar.

higher in the thrombolytic patients, the rate of the more relevant and objective end points of amputation and death were equivalent (Fig. 2). Articles comprising subgroup analyses of the STILE data appeared, one relating to native artery occlusions33 and one to bypass graft occlusions.34 Thrombolysis appeared more effective in patients with graft occlusions. The rate of major amputation was higher in native arterial occlusions treated with thrombolysis (10% thrombolysis vs. 0% surgery at 1 year; p = 0.0024). By contrast, amputation was lower in patients with acute graft occlusions treated with thrombolysis (p = 0.026). These data suggest that thrombolysis may be of greatest benefit in patients with acute bypass graft occlusions of less than 14 days.

Table II. Results in TOPAS trial of recombinant urokinase versus surgery for acute peripheral arterial occlusion Urokinase group Surgery group (n = 272) (n = 272)

Urokinase group Surgery group (n = 272) (n = 272)

Operative intervention

6 months 1 year 6 months 1 year (n) (n) (n) (n) Worst Outcomea

6 months 1 year 6 months 1 year (%) (%) (%) (%)

Amputation Above the knee Below the knee Open surgical procedures Major Moderate Minor Percutaneous procedures

48 22 26 315 102 89 124 128

16 12.2 5.6 6.6 40.3 23.6 10.3 6.4 16.9 14.6

a

58 25 33 351 116 98 137 135

41 19 22 551 177 136 238 55

51 26 25 590 193 145 252 70

Death Amputation Above the knee Below the knee Open surgical procedures Major Moderate Minor Endovascular procedures Medical treatment alone

20 15 6.5 8.5 39.3 24.3 8.7 6.3 15.4 10.3

12.3 12.9 6.1 6.8 69 39.3 16.3 13.4 2.1 3.7

17 13.1 7.5 56 65.4 39.3 13.4 12.7 1.7 2.8

Worst outcome is the most severe event that occurred over the specified time period. Values given are percentage of patients.

Vol. 16, No. 6, 2002

The third and final randomized comparison of thrombolysis and surgery was the Thrombolysis Or Peripheral Arterial Surgery (TOPAS) trial, funded by Abbott Laboratories. Following completion of a preliminary dose-ranging trial in 213 patients,35 544 patients were randomized to a recombinant form of UK or primary operative intervention.6 After a mean follow-up period of 1 year, the rate of amputation-free survival was identical in the two treatment groups—68.2% and 68.8% in the UK and surgical patients, respectively (Table II). While this trial failed to document improvement in survival or limb salvage with thrombolysis, fully 31.5% of the thrombolytic patients were alive without amputation with nothing more than a percutaneous procedure after 6 months of follow-up. After 1 year, this number had decreased only slightly, with 25.7% alive, without amputation, and with only percutaneous interventions. Thus, the original goal of the TOPAS trial, to generate data on which regulatory approval of recombinant UK would be based, was not achieved. Nevertheless, the findings confirmed that acute limb ischemia could be managed with catheter-directed thrombolysis, achieving similar amputation and mortality rates but avoiding the need for open surgical procedures in a significant percentage of patients.

Thrombolysis for peripheral arterial occlusive 803

9. 10.

11.

12.

13.

14.

15.

16.

17.

18.

REFERENCES 19. 1. Blaisdell FW, Steele M, Allen RE. Management of acute lower extremity arterial ischemia due to embolism and thrombosis. Surgery 1978;84:822-834. 2. Jivega˚ rd L, Holm J, Scherste´ n T. Acute limb ischemia due to arterial embolism or thrombosis: influence of limb ischemia versus pre-existing cardiac disease on postoperative mortality rate. J Cardiovasc Surg 1988;29:32-36. 3. Holm J, Scherste´ n T. Anticoagulant treatment during and after embolectomy. Acta Chir Scand 1972;138(683): 687. 4. Caruana JA, Gutierrez IZ, Andersen MN, et al. Factors that affect the outcome of peripheral arterial embolization. Arch Surg 1981;116:423-425. 5. 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. 6. Ouriel K, Veith FJ, Sasahara 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. 7. Anonymous. Results of a prospective randomized trial evaluating surgery versus thrombolysis for ischemia of the lower extremity. The STILE trial. Ann Surg discussion 1994;220:251-266. 8. Ouriel K, Kandarpa K, Schuerr DM, Hultquist M, Hodkinson B, Wallin B. Prourokinase versus urokinase for recanaliza-

20.

21.

22. 23.

24. 25.

26.

27.

tion of peripheral occlusions, safety and efficacy: the PURPOSE trial. J Vasc Interv Radiol 1999;10:1083-1091. Dormandy J, Heeck L, Vig S. Acute limb ischemia. Semin vasc Surg 1999;12:148-153. McNamara TO, Fischer JR. Thrombolysis of peripheral arterial and graft occlusions: improved results using highdose urokinase. AJR Am J Roentgenol 1985;144:769775. Semba CP, Murphy TP, Bakal CW, Calis KA, Matalon TA. Thrombolytic therapy with use of alteplase (rt-PA) in peripheral arterial occlusive disease: review of the clinical literature. The Advisory Panel. J Vasc Interv Radiol 2 Pt 1 2000;11:149-161. Ouriel K, Katzen B, Mewissen MW, et al. Reteplase in the treatment of peripheral arterial and venous occlusion. J Vasc Interv Radiol 2000;11:849-854. McNamara TO. Thrombolysis as the initial treatment for acute lower limb ischemia. In: Comerota AJ ed. J.B. Lippincott, Philadelphia, 1997pp 253-268. Faggioli GL, Peer RM, Pedrini L, et al. Failure of thrombolytic therapy to improve long-term vascular patency. J Vasc Surg 1994;19:289-296. Korn P, Khilnani NM, Fellers JC, et al. Thrombolysis for native arterial occlusions of the lower extremities: clinical outcome and cost. J Vasc Surg 2001;33:1148-1157. Sullivan KL, Gardiner GAJ, Kandarpa K, et al. Efficacy of thrombolysis in infrainguinal bypass grafts. Circulation 1991;83 (2 Suppl):99-105. Milstone H. A factor in normal human blood which participates in streptococcal fibrinolysis. J Immunol 42:116. Onundarson PT, Haraldsson HM, Bergmann L, Francis CW, Marder VJ. Plasminogen depletion during streptokinase treatment or two-chain urokinase incubation correlates with decreased clot lysability ex vivo and in vitro. Thromb Haemost 1993;70:998-1004. Tillett WS, Garner RL. The fibrinolytic activity of hemolytic streptococci. J Exp Med 1933;58:485. Tillett WS, Sherry S. The effect in patients of streptococcal fibrinolysin (streptokinase) and streptococcal desoxyribonuclease on fibrinous, purulent, and sanguinous pleural exudations. J Clin Invest 1949;28:173. Tillett WS, Johnson AJ, McCarty WR. The intravenous infusion of the streptococcal fibrinolytic principle (streptokinase) into patients. J Clin Invest 1955;34:169-185. Cliffton EE. The use of plasmin in humans. Ann NY Acad Sci 1957;68:209-229. Jostring H, Barth U, Naidu R. Changes of antistreptokinase titer following long-term streptokinase therapy. In: Martin M, Schoop W, Hirsh J eds. Hans Huber, Vienna, 1978pp 110. Macfarlane RG, Pinot JJ. Fibrinolytic activity of normal urine. Nature 1947;159:779. Sobel GW, Mohler SR, Jones NW, Dowdy ABC, Guest MM. Urokinase: an activator of plasma fibrinolysin extracted from urine. Am J Physiol 1952;171:768-769. Credo RB, Burke SE, Barker WM, et al. Recombinant urokinase (r-UK): biochemistry, pharmacology, and clinical experience. In: Sasahara AA, Loscalzo J eds. Marcel Dekker, New York, 1978 pp 513-537. Husain SS, Lipinski B, Gurewich V, et al. (1979) Isolation of plasminogen activators useful as therapeutic and diagnostic agents (single-chain, high-fibrin affinity urokinase). U.S. Patent # 4,381, 346.

804 Ouriel

28. Hoylaerts M, Rijken DC, Lijnen HR, Collen D. Kinetics of the activation of plasminogen by human tissue plasminogen activator: role of fibrin. J Biol Chem 1982;257: 2912. 29. The Gusto Investigators.. An International randomized trial comparing four thrombolytic therapies for acute myocardial infarction. N Engl J Med 1993;329:673-682. 30. Cannon CP, Gibson CM, McCabe CH, et al. TNK-tissue plasminogen activator compared with front-loaded alteplase in acute myocardial infarction: results of the TIMI 10B trial. Thrombolysis in Myocardial Infarction (TIMI) 10B Investigators. Circulation 1998;98:2805-2814. 31. Meierhenrich R, Carlsson J, Seifried E, et al. Effect of reteplase on hemostasis variables: analysis of fibrin specificity, relation to bleeding complications and coronary patency. Int J Cardiol 1998;65:57-63.

Annals of Vascular Surgery

32. Valji K. Evolving strategies for thrombolytic therapy of peripheral vascular occlusion. J Vasc Inter Radiol 2000;11:411420. 33. Weaver FA, Comerota AJ, Youngblood M, Froehlich J, Hosking JD, Papanicolaou G. Surgical revascularization versus thrombolysis for nonembolic lower extremity native artery occlusions: results of a prospective randomized trial. The STILE Investigators. Surgery versus thrombolysis for ischemia of the lower extremity. J Vasc Surg 1996;24:513521. 34. Comerota AJ, Weaver FA, Hosking JD, et al. Results of a prospective, randomized trial of surgery versus thrombolysis for occluded lower extremity bypass grafts. Am J Surg 1996;172:105-112. 35. Ouriel K, Veith FJ, Sasahara AA. Thrombolysis or peripheral arterial surgery: phase I results. TOPAS Investigators. J Vasc Surg 1996;23:64-73.