Tissue Plasminogen Activator for the Treatment of Deep Venous Thrombosis of the Lower Extremity* Alan Forster, MD; and Philip Wells, MD

Transcription

1 review Tissue Plasminogen Activator for the Treatment of Deep Venous Thrombosis of the Lower Extremity* A Systematic Review Alan Forster, MD; and Philip Wells, MD Objective: To assess, by systematic review, the efficacy and safety of recombinant tissue plasminogen activator (rt-pa) in the treatment of lower extremity deep venous thrombosis (DVT). A secondary objective is to assess the optimal dose and route of administration of rt-pa. Methods: Included studies were randomized, controlled trials comparing rt-pa plus unfractionated heparin (UFH) to UFH alone, rt-pa at different doses, or rt-pa by different routes of administration in the treatment of DVT. Outcomes had to be described in terms of percent change in venographic patency for efficacy (>50% lysis) and in sufficient detail for complications (major and minor hemorrhages and other). The search strategy included searching electronic databases and contacting pharmaceutical agencies and content experts. Study quality was assessed using the Jadad scale. A threshold quality score was used to exclude trials. Results: Five studies met the following inclusion criteria: three comparing rt-pa plus UFH vs UFH alone (180 patients); one comparing high-dose vs low-dose rt-pa (32 patients); and one comparing systemic vs local administration of rt-pa (151 patients). In studies comparing rt-pa vs placebo, patients assigned to rt-pa were more likely to have > 50% lysis and complications (summary odds ratios [OR], 11.7; 95% confidence interval [CI], 2.61 to 52.5; and OR, 9.95; 95% CI, 2.21 to 44.7, respectively). Major and intracerebral hemorrhages were not significantly increased. One study comparing different doses demonstrated that high-dose and low-dose rt-pa were equally efficacious (OR, 0.88; 95% CI, 0.05 to 14.78). Local rt-pa was neither more efficacious nor riskier than systemic rt-pa. Conclusion: This systematic review does not support routine use of rt-pa for DVT. (CHEST 2001; 119: ) Key words: systematic review; tissue plasminogen activator; venous thromboembolism Abbreviations: CI confidence interval; DVT deep venous thrombosis; OR odds ratio; rt-pa recombinant tissue plasminogen activator; SK streptokinase; UFH unfractionated heparin Deep venous thrombosis (DVT) of the lower extremity is a common condition. 1 It causes significant pain in the acute setting and is associated *From the Divisions of General Internal Medicine (Dr. Forster) and Hematology (Dr. Wells), Ottawa Hospital, Civic Campus, University of Ottawa, and the Loeb Research Institute, Clinical Epidemiology Unit (Dr. Wells), Ottawa, Ontario, Canada. Dr. Wells is a recipient of a research scholarship from the Heart and Stroke Foundation of Canada. Manuscript received November 23, 1999; revision accepted June 14, Correspondence to: P. Wells, MD, Division of Hematology, 737 Parkdale Ave, Suite 452, Ottawa, Ontario, K1Y 1J8, Canada; pwells@ottawahospital.on.ca with the life-threatening illness, pulmonary embolism. 1,2 In rare cases, it can lead to limb-threatening ischemia. Also, it is often associated with the longterm problem of postphlebitic syndrome, which may be characterized by pain, edema, skin discoloration, or skin ulcers. These long-term complications are the result of chronic venous obstruction and irreversible damage to the venous valves. 2,3 Thrombolytic agents, by actively dissolving the thrombus, could treat cases in which life-threatening limb ischemia exists, as well as potentially prevent postphlebitic syndrome. These potential benefits must be weighed against the potential for an in- 572 Reviews

2 College of Chest Physicians on antithrombotic therapy concluded that the lack of evidence indicates that the use of thrombolytic agents should be highly individualized. No firm recommendations were made. The purpose of this review is to provide a summary of the evidence of effectiveness of this agent for DVT. It will assess the efficacy (short-term venographic patency) and safety profiles (major and minor hemorrhage, pulmonary embolism) of rt-pa plus heparin vs heparin alone in the treatment of lower extremity DVT. It will also assess whether the dose or route of administration of rt-pa is related to outcomes. Materials and Methods Search Strategy Electronic databases were searched using the OVID search engine for MEDLINE (1966 through September 1999) and Current Contents (1997 through week 33, 1999). The search terms are listed in the Appendix. We contacted the authors of the published randomized trials that we included in the review and the pharmaceutical companies that manufacture rt-pa to determine whether there were any unpublished studies. Bibliographies of identified articles were searched to identify further articles. One reviewer (AF) performed the search. Both authors reviewed included studies for quality assessment Figure 1. Trial flow. creased risk of hemorrhage. There are also anecdotal reports of pulmonary embolism in DVT patients receiving thrombolytic agents. 4 There are a number of thrombolytic agents. Streptokinase (SK) is the most widely studied. Several systematic reviews 3,5 and a decision analysis 6 comparing SK vs unfractionated heparin (UFH) in the treatment of DVT have been published. The evidence suggests that SK results in higher rates of vein patency and hemorrhagic complications. 5 The decision analysis concludes that short-term risks associated with SK are unacceptable to patients with DVT. 6 Tissue plasminogen activator and its analog, recombinant tissue plasminogen activator (rt-pa), theoretically have several advantages over SK. rt-pa is thrombus-specific and, therefore, could result in fibrinolysis only at the clot site rather than systemically. In addition, it has a shorter half-life and should therefore have a lower risk associated with it. 7 Several clinical trials have assessed the use of rt-pa for DVT, yet there is no conclusive evidence in the literature with respect to its benefit or its risk. The recent consensus conference of the American Inclusion/Exclusion Criteria Studies selected for the review had to be randomized, controlled trials. The following three types of interventions were assessed: rt-pa plus heparin vs heparin alone; high-dose rt-pa vs low-dose rt-pa; and locally administered rt-pa vs systemic rt-pa. Only studies evaluating patients with lower leg DVTs were included. Efficacy had to be described in terms of the percentage of thrombus lysed to be included in the review. Hemorrhagic and other adverse events had to be described in detail. Studies without a comparison group (case series) were excluded from the review. Studies scoring 3 on the Jadad scale 8 for trial quality were not included. The Jadad scale is a validated measurement tool assessing the internal validity of a randomized trial. Studies scoring 3 are considered to be of low quality. 8 Data Abstraction Both authors reviewed included studies. Data from the trials were abstracted using standardized forms. End Points This review used the following end points: 50% lysis of thrombus; all complications (see below); and major hemorrhage and intracerebral hemorrhage. The extent of lysis was determined by comparing the results of pretreatment and posttreatment venography. Major hemorrhage includes a drop in hemoglobin 2 g/dl, hemorrhage requiring blood transfusion, or clinically overt intracerebral, intraarticular, retroperitoneal, in- CHEST / 119 / 2/ FEBRUARY,

3 Table 1 Excluded Studies* Title Author Publication Year Reason for Exclusion A strategy of aggressive regional therapy for acute iliofemoral venous thrombosis with contemporary venous thrombectomy of catheter directed thrombolysis Fibrinolytic therapy of deep leg- and pelvic vein thromboses with various dosages of rt-pa: a pilot study Comerata et al Case series Doder et al Included patients with PE. Did not adequately describe outcomes. Garcia-Rubi et al Case series Treatment of acute peripheral thromboembolic disease with the new thrombolytic agent rt-pa Comparison of results of fibrinolytic therapy with urokinase, Grimm et al No randomization ultra-high dose streptokinase and recombinant tissue-type plasminogen activator in deep vein thrombosis Lysis block technique: a new method of fibrinolytic treatment Heimig et al Case series with tissue-type plasminogen activator (rt-pa) Comparison of dosage schedules of rt-pa in the treatment of Marder et al Nonrandomized design. proximal deep vein thrombosis Short term response to thrombolytic therapy in deep venous Meyerovitz et al Results published in an included thrombosis: predictive value of venographic appearance study Frequent asymptomatic pulmonary embolism in patients with Moser et al Venographic outcomes not deep venous thrombosis reported nonrandomized design Loco-regional thrombolysis in deep venous thrombosis Palombo et al Case series Pilot trial of recombinant tissue type plasminogen activator in Pilger et al Case series deep venous thrombosis Loco-regional thrombolysis therapy of deep venous thromboses Rudofsky and 1993 No randomization Timmermann 19 Specific lysis of an iliofemoral thrombus by administration of Weimar et al Case report extrinsic (tissue-type) plasminogen activator Fibrinolytic therapy of deep vein thrombosis with continuous Zimmerman et al Nonrandomized design intravenous infusion of a recombinant tissue plasminogen activator rt-pa thrombolysis of deep vein thrombosis with different Zimmerman et al Insufficient data to include dosages: results of a prospective multicenter study Throbolysetherapie der tiefen venosen thombose mit rt-pa Zimmerman et al Case series *PE pulmonary embolism. traocular, or GI hemorrhage. All complications included allergic reactions, major and minor hemorrhage (which includes bleeding at venipuncture sites, mucosal bleeding, superficial hematoma, or other hemorrhage not meeting the requirements to be classified as a major hemorrhage), and any other complication necessitating the discontinuation of therapy. Statistical Methods Mantel-Haenszel odds ratios (ORs) are used to estimate the summary (OR) and their respective 95% confidence intervals (CIs). Results of Search Results Figure 1 illustrates the flow of studies through this review. Six hundred forty potentially relevant studies were identified by the electronic database search. Content experts did not identify unpublished studies. Only one pharmaceutical agency responded to our request. Hoffman LaRoche identified two additional studies not identified by our database search because they also searched Embase. A reviewing of the bibliographies of retrieved articles identified two further studies. Thus, there were 644 potentially relevant studies identified by the search. Of the 644 potentially relevant studies, 261 were excluded because the title of the article did not address the question posed by the review. The abstracts for the remaining 383 studies were screened. A further 363 articles were excluded as they did not address the specific question or were review articles. The remaining 20 studies were retrieved. Fifteen of these studies were excluded because they did not meet the inclusion criteria required. The excluded studies and the reasons for exclusion are listed in Table 1. No studies were excluded because of a score of 3 on the Jadad scale. Characteristics of Included Studies The five included studies had very different study designs. These differences are captured in the data 574 Reviews

4 Table 2 Description of Included Studies Outcomes Protocol/Author Patients, No. Inclusion Criteria Intervention Efficacy (% lysis) Bleeding (Major/Minor) Quality Score* rt-pa plus UFH vs UFH alone Turpie et al, Diagnosis within 7dof symptoms Turpie et al, Diagnosis within 7dof symptoms Goldhaber et al, Diagnosis within 14 d of symptoms Verhaeghe et al, rt-pa: High dose vs low dose Bounameaux et al, Diagnosis within 10 d of symptoms 32 Diagnosis within 7dof symptoms rt-pa: Systemic vs local administration Schwieder et al, Diagnosis within 14 d of symptoms Group 1: injection of 0.5 mg/kg twochain rt-pa for 4 h Yes Yes 3 Group 2: injection of placebo Both groups received continuous IV UFH Group 1: injection of 0.5 mg/kg Yes Yes 3 single-chain rt-pa for 8 h, repeated 1at24h Group 2: injection of placebo for 8 h repeated 1at24h Both groups received continuous IV UFH Continuous infusion of rt-pa for 1 d Yes Yes 3 Group 1: 0.05 mg/kg/h Group 2: 0.05 mg/kg/h UFH Group 3: placebo UFH Group 1: 2dofcontinuous infusion of rt-pa: day 1, 100 mg; day 2, 50 mg. Group 2: 48 h continuous infusion of rt-pa: both days 50 mg Group 3: 2dofcontinuous infusion of placebo All groups received IV UFH Continuous infusion of different doses rt-pa for 7 d Group 1: 0.5 mg/kg/d Group 2: 0.25 mg/kg/d Both groups received IV UFH 4 7 daily injections of 20 mg of rt- PA for 4 h Group 1: systemic administration Group 2: local administration No Yes 4 Yes Yes 5 Yes Yes 3 *Quality is measured using the Jadad scale. This is a measurement tool for evaluating a trial s internal consistency by assessment of adequacy of randomization, double blinding, and description of withdrawals and dropouts. Trials can score up to 5 points, and 3 points reflects moderate to high quality. The study by Turpie et al 25 has been listed twice to reflect the two separate protocols described in the single publication. presented in Table 2. In summary, three of the studies assessed rt-pa plus heparin vs heparin alone, one study assessed rt-pa by local administration vs by systemic administration, and one study compared two different doses of rt-pa. The studies by Goldhaber et al 24 (65 patients), Turpie et al 25 (83 patients), and Verhaeghe et al 26 (21 patients) enrolled a total of 169 patients comparing rt-pa plus UFH or placebo with UFH alone. Important differences in inclusion criteria, method of administration of rt-pa, and outcome assessment existed among these studies. Patients enrolled had symptoms present for up to 14 days (Goldhaber et al 24 ), 10 days (Verhaeghe et al 26 ), or 7 days (Turpie et al 25 ). Three different methods of administration were used. Goldhaber et al 24 administered the rt-pa in a continuous infusion for 24 h; Verhaeghe et al 26 gave a bolus then ran a continuous infusion for the CHEST / 119 / 2/ FEBRUARY,

5 Table 3 rt-pa vs Placebo* Study rt-pa Control OR (95% CI) Efficacy Goldhaber et al 24 15/53 (28.3) 0/12 (0) 10.1 ( ) Turpie et al (2) 25 6/29 (20.7) 2/30 (6.7) 3.65 ( ) Turpie et al (1) 25 7/12 (58.3) 0/12 (0) 34.1 ( ) Total 28/94 (29.8) 2/54 (3.7) 11.7 ( ) Complications (all) Goldhaber et al 24 12/53 (22.6) 0/12 (0) 7.53 ( ) Turpie et al (2) 25 1/29 (3.4) 1/30 (3.3) 1.04 ( ) Turpie et al (1) 25 4/12 (33.3) 1/12 (8.3) 5.50 ( ) Verhaeghe et al 26 16/25 (64.0) 0/7 (0) 26.1 ( ) Total 33/119 (27.7) 2/61 (3.3) 9.95 ( ) Complications (major hemorrhage) Goldhaber et al 24 1/53 (1.89) 0/12 (0) 0.71 ( ) Turpie et al (2) 25 0/29 (0) 1/30 (3.3) 0.33 ( ) Turpie et al (1) 25 1/12 (8.3) 1/12 (8.3) 1.00 ( ) Verhaeghe et al 26 8/25 (32) 0/7 (0) 7.29 ( ) Total 10/119 (8.4) 2/61 (3.3) 2.52 ( ) Complications (intracranial hemorrhage) Goldhaber et al 24 1/53 (1.89) 0/12 (0) 0.71 ( ) Turpie et al (2) 25 0/29 (0) 0/30 (0) NE Turpie et al (1) 25 0/12 (0) 0/12 (0) NE Verhaeghe et al 26 0/25 (0) 0/7 (0) NE Total 1/119 (0.8) 0/61 (0) 1.55 ( ) *Values given as No. of events/no. of patients (%), unless otherwise indicated. Treatment was efficacious if there was 50% lysis on the posttreatment venogram. The trial by Verhaeghe et al 26 is not included because it did not report efficacy in terms of percentage of thrombus lysed. Therefore, it could not be included in the summary estimate of efficacy. The results of the trial are included for the complications as sufficient description was included in the manuscript. next 48 h; and Turpie et al 25 administered a bolus of rt-pa (in phase 1 of the study, one dose was given for 4 h and in phase 2, two doses were given, each for 8 h). Outcome assessment was based on comparison of pretreatment and posttreatment venograms in all three studies (posttreatment venography was performed within 24 h of rt-pa completion in the three studies). However, because Verhaeghe et al 26 reported change in the Marder score, whereas Turpie et al 25 and Goldhaber et al 24 presented the percentage of lysis, the study by Verhaeghe et al 26 could not be included in the analysis of efficacy. All three studies had adequate descriptions of adverse outcomes to allow pooling of results, therefore, the study by Verhaeghe et al 26 is included in the analysis of adverse event rate. Bounameaux et al 27 compared two different doses of rt-pa in 32 patients who had symptoms for 7 days. All patients received UFH by continuous IV infusion but were randomized to receive one of two doses of rt-pa (0.25 mg/kg/h or 0.5 mg/kg/h) by continuous infusion for 7 days or until clinical improvement. Patients and physicians were blinded to treatment group. Outcomes were assessed by comparison of posttreatment venography and were classified into grades of improvement. Schwieder et al 28 enrolled 151 patients in their comparison of the two different methods of administering rt-pa. All patients received UFH by continuous IV infusion. There was no placebo group. Schwieder et al 28 included patients with symptoms present for up to 2 weeks. Participants were randomized to daily injections of rt-pa, 20 mg, given into a pedal vein of the symptomatic leg or into a brachial vein. Injections were given for 4 h and repeated daily for 7 days or until clinical improvement. Clinical assessors were not blinded to treatment group. Outcome was measured at the percentage improvement in the posttreatment venogram. Summary of Outcomes Table 3 demonstrates the individual trial results and the summary result comparing rt-pa plus heparin vs heparin alone. Patients in the treatment group were more likely to have both 50% lysis and adverse events. The summary ORs are 11.7 (95% CI, 2.61 to 52.5) and 9.95 (95% CI, 2.21 to 44.7), respectively. Major hemorrhage and intracerebral hemorrhage were not significantly increased, although there was a trend toward increased risk (for major hemorrhage, OR, 2.52; 95% CI, 0.48 to 13.29; and for intracerebral hemorrhage, OR, 1.55; 95% CI, 0.03 to 989.8). In addition, two of the major hemorrhages were in patients who had undergone recent 576 Reviews

6 Table 4 rt-pa vs Placebo: Absolute Risk Difference, Number Needed to Treat, and Number Needed to Harm* Outcome Absolute Risk Difference, % Event Rate in Control Group, % NNT NNH Clot lysis All complications Major hemorrhage ICH *ICH intracerebral hemorrhage; NNT number needed to treat (ie, the No. of people people with lower extremity DVT who need to be treated with rt-pa to see one with 50% lysis); NNH number needed to harm (ie, the number of people with lower extremity DVT who will be treated with rt-pa before there is an adverse event). major orthopedic surgeries. There was only one intracerebral hemorrhage in the 119 patients randomized to receive rt-pa plus heparin. Table 4 demonstrates the absolute risk difference and the event rate in the control group, as well as the numbers needed to treat and to harm. This table demonstrates that four patients required treatment with rt-pa to achieve success, 50% lysis, in one patient. However, for every 5 patients treated, there was one additional complication, and for every 15 patients, one major hemorrhage. Table 5 demonstrates a summary of the results of rt-pa administered for 7 days at a dose of 0.5 mg/kg/d vs a dose of 0.25 mg/kg/d. There was no benefit in terms of efficacy at the higher dose. Although there was a trend toward increased risk for all adverse events, none of the estimates was statistically significant. Table 6 illustrates that rt-pa given locally is no more effective than that given systemically. In addition, there is a trend toward an increase in the number of adverse events. Discussion This systematic review does not support routine use of rt-pa for DVT of the lower extremities. Although increased venographic patency occurs, the increased risk of complications is such that a difference in important clinical outcomes would be necessary to change practice. In addition, higher doses of rt-pa do not appear to be more efficacious and are possibly more dangerous than lower doses. Locally administered rt-pa may have more complications and is not any more effective than systemic administration. These conclusions must be qualified by several limitations inherent in the study designs of the included articles and also in the systematic review process. The clinical importance of the surrogate outcomes, venographic patency, and all complications, is not known. Whether having 50% lysis of a thrombus at 1 week makes any improvement in subsequent development of postphlebitic syndrome is not known. Evidence that it might is provided in the study by Turpie et al. 25 In their cohort, patients with 50% lysis were less likely to experience symptoms during long-term follow-up. The magnitude of this difference needs further exploration in future studies. Additionally, the clinical meaning of all complications is uncertain. Minor hemorrhage, which makes up by far the greatest number of these, is often clinically irrelevant. The lack of power to detect an increased risk of major complications is also a major drawback of the review. The major complications are certainly clinically relevant; however, none of the summary measures were statistically significant. This could be a reflection of the low numbers of patients studied, as evidenced by the wide confidence intervals. Alternatively, it could mean that major complications are truly not increased in these patients. Before giving up on this potentially efficacious therapy, we need to know more about the risks. The inclusion of patients with a low probability of responding to therapy decreases the ability of any Table 5 High-Dose rt-pa vs Low-Dose rt-pa* Study High Dose Low Dose OR (95% CI) Efficacy 1/17 (5.9) 1/15 (6.7) 0.88 ( ) Complications All 11/17 (64.7) 4/15 (26.7) 5.04 ( ) Major hemorrhage 5/17 (29.4) 2/15 (13.3) 2.71 ( ) ICH 0/17 (0) 0/15 (0) NE *Values given as No. of events/no. of patients (%), unless otherwise indicated. See Tables 3 and 4 for other abbreviations not used in text. CHEST / 119 / 2/ FEBRUARY,

7 Table 6 Local rt-pa vs Systemic rt-pa Study Local Systemic OR (95% CI) Efficacy 21/79 (26.6) 22/72 (30.6) 0.82 ( ) Complications All 26/79 (32.9) 16/72 (22.2) 1.71 ( ) Major hemorrhage 8/79 (10.1) 4/72 (5.6) 1.91 ( ) ICH 0/79 (0) 1/72 (1.4) 0.30 ( ) *Values given as No. of events/no. of patients (%), unless otherwise indicated. See Table 4 for abbreviation not used in text. study to demonstrate a benefit. Several studies included patients who had had symptoms for more than a week. These older clots may be less amenable to lysis owing to increased fibrin cross-linking. Inclusion of such patients diminishes the summary estimate of effectiveness. The inclusion of patients with a high probability of experiencing an adverse event likewise biases against the therapy. Two episodes of major hemorrhage occurred in patients in whom DVT developed secondary to joint replacement therapy. Postoperative patients should not receive thrombolytic agents for DVT. Also, several studies included patients 65 years of age. As hemorrhage risk increases with age, inclusion of older patients will again bias the result of the study. 29 All systematic reviews are limited by the possibility of publication bias. An unpublished trial with many major complications may change the conclusion significantly. This study minimized this bias by attempting to contact investigators and pharmaceutical companies. Systematic reviews are also limited by being retrospective in nature. Thus, the lack of long-term outcomes in the included studies limits the conclusion. Additionally, the different interventions assessed in the trials (doses, continuous vs bolus doses, duration of therapy, and types of rt-pa) make it impossible to recommend a single approach. In summary, we believe that the weight of evidence does not support routine use of rt-pa in patients with lower extremity DVT. However, there is insufficient evidence regarding its risks to discard this potentially effective treatment in all patients. Patients with limb-threatening thrombosis and a high risk of severe postphlebitic syndrome could benefit if the risk of bleeding was not excessive. However, this needs to be evaluated in a randomized trial. The rarity of this clinical scenario will necessitate that such a trial be multicentered. Appendix MeSH Headings Used in Electronic Database Search 1. clinical trial.pt. 2. randomized controlled trial.pt. 3. tu.fs. 4. dt.fs. 5. random$.tw. 6. (double adj blind$).tw. 7. placebo$.tw. 8. or/ deep venous thrombosi.mp. 10. *thrombophlebitis 11. or/ exp streptokinase/ or streptokinas.mp. 13. exp tissue plasminogen activator/or tissue plasminogen activator.mp. 14. fibrinolytic agents/ad 15. fibrinolytic agents/ae 16. *thrombolytic therapy/ 17. or/ and 11 and 17 References 1 Kelley W. Textbook of internal medicine. 2nd ed. Philadelphia, PA: JB Lippincott, 1992; Strandness D, Langlois Y, Cramer M, et al. Long term sequelae of acute venous thrombosis. JAMA 1983; 250: Francis C, Marder V. Fibrinolytic therapy for venous thrombosis. Prog Cardvasc Dis 1991; 34: Holmstrom M, Bratt G Tormebohm E, et al. Fatal pulmonary embolism caused by streptokinase treatment of deep venous thrombosis of the leg? J Intern Med 1990; 228: Rogers L, Lutcher C. Streptokinase therapy for deep vein thrombosis: a comprehensive review of the English literature. Am J Med 1990; 88: O Meara J, McNutt R, Evans A, et al. A decision analysis of streptokinase plus heparin as compared with heparin alone for deep vein thrombosis. N Engl J Med 1994; 330: Hyers TM, Agnelli G, Hull RD, et al. Antithrombotic therapy for venous thromboembolic disease: fifth ACCP consensus conference on antithrombotic therapy. Chest 1998; 114(suppl): 561S 578S 8 Jadad AR, Moore RA, Carroll C, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996; 17: Comerata AJ, Aldridge SC, Cohen G, et al. A strategy of aggressive regional therapy for acute iliofemoral venous thrombosis with contemporary venous thrombectomy of catheter directed thrombolysis. J Vasc Surg 1994; 20: Doder A, Pilger E, Obernosterer A, et al. Fibrinolytic therapy of deep leg- and pelvic vein thrombosis with various dosages of rt-pa: a pilot study. Vasa Suppl 1992; 37: Garcia-Rubi DE, Henkel RD, Tejeda RH, et al. Treatment of acute peripheral thromboembolic disease with the new thrombolytic agent rt-pa (tissue activator of recombinant plasminogen). Invest Med Int 1991; 18: Reviews

8 12 Grimm W, Schwieder G, Wagner T. Comparison of results of fibrinolytic therapy with urokinase, ultra-high dose streptokinase and recombinant tissue-type plasminogen activator in deep vein thrombosis. Blood 1990; 60: Heimig T, Martin M, Fiebach BJO, et al. Lysis block technique: a new method of fibrinolytic treatment with tissue-type plasminogen activator (rt-pa). Vasa Suppl 1993; 39: Marder V, Brenner B, Totterman S, et al. Comparison of dosage schedules of rt-pa in the treatment of proximal deep vein thrombosis. J Lab Clin Med 1992; 119: Meyerovitz M, Polak J, Goldhaber S. Short term response to thrombolytic therapy in deep venous thrombosis: predictive value of venographic appearance. Radiology 1992; 184: Moser K, Fedullo P, LitteJohn J, et al. Frequent asymptomatic pulmonary embolism in patients with deep venous thrombosis. JAMA 1994; 271: Palombo D, Porta C, Brustia P, et al. Loco-regional thrombolysis in deep venous thrombosis. Phlebologie 1993; 46: Pilger E, Obernosterer A, Decrinis M, et al. Pilot trial of recombinant tissue type plasminogen activator in deep venous thrombosis. Fibrinolysis 1992; 6: Rudofsky G, Timmermann J. Loco-regional thrombolysis therapy of deep venous thromboses. Z Kardiol 1993; 82(suppl): Weimar W, Stibbe J, van Seyen AJ, et al. Specific lysis of an iliofemoral thrombus by administration of extrinsic (tissuetype) plasminogen activator. Lancet 1981; 2: Zimmerman R, Gursoy A, Horn A, et al. Fibrinolytic therapy of deep vein thrombosis with continuous intravenous infusion of a recombinant tissue plasminogen activator. Semin Thromb Hemost 1991; 17: Zimmerman R, Seifried E, Schramm W, et al. rt-pa thrombolysis of deep vein thrombosis with different dosages: results of a prospective multicenter study. Paper presented at: Thirteenth Congress of the International Society on Thrombosis and Haemostasis, Amsterdam, 1991; Zimmerman R, Horn A, Harenberg H, et al. Thrombolystherapie der tiefen venosen thrombose mit rt-pa. Klin Wochenschr 1988; 66(suppl): Goldhaber S, Meyerovitz M, Green D, et al. Randomized controlled trial of tissue plasminogen activator in proximal deep venous thrombosis. Am J Med 1990; 88: Turpie A, Levine M, Hirsh J, et al. Tissue plasminogen activator (rt-pa) vs heparin in deep vein thrombosis: results of a randomized trial. Chest 1990; 97(suppl):172S 175S 26 Verhaeghe R, Besse P, Bounameaux H, et al. Multicenter pilot study of the efficacy and safety of systemic rt-pa administration in the treatment of deep vein thrombosis of the lower extremities and/or pelvis. Thromb Res 1989; 55: Bounameaux H, Banga J, Bluhmki E, et al. Double-blind, randomized comparison sf systemic continuous infusion of 0.25 mg vs 0.50 mg/kg/24h of alteplase over 3 to 7 days for treatment of deep venous thrombosis in heparinized patients: results of the European Thrombolysis with rt-pa in Venous Thrombosis (ETTT) Trial. Thromb Haemost 1992; 67: Schwieder G, Grimm W, Siemens H, et al. Intermittent regional therapy with rt-pa is not superior to systemic thrombolysis in deep vein thrombosis (DVT): a German multicenter trial. Thromb Haemost 1995; 74: Holmes D, Califf R, Topol E. Lessons we have learned from the Gusto Trial. J Am Coll Cardiol 1995; 25(suppl):10S 17S CHEST / 119 / 2/ FEBRUARY,