J. Harenberg 1, J. A. Schmidt 2, K. Koppenhagen 3, A. Tolle 4, M. V. Huisman 5, H. R. Büller 6 and the EASTERN Investigators*

Transcription

1 2000 Schattauer Verlag, Stuttgart Fixed-dose, Body Weight-independent Subcutaneous LMW Heparin versus Adjusted Dose Unfractionated Intravenous Heparin in the Initial Treatment of Proximal Venous Thrombosis J. Harenberg 1, J. A. Schmidt 2, K. Koppenhagen 3, A. Tolle 4, M. V. Huisman 5, H. R. Büller 6 and the EASTERN Investigators* From the 1 1. Dept. of Medicine, University Hospital, Mannheim, Germany, 3 Dept. of Radiology, Benjamin Franklin University, Berlin, Germany, 2 Dept. of Angiology, Otto-von-Guericke-University, Magdeburg, Germany, 4 Novartis Pharma GmbH, Nuremberg, Germany; 5 Dept. of General Internal Medicine, University Hospital, Leiden, Netherlands, 6 Dept. of Vascular Medicine, Academic Medical Center, Amsterdam, Netherlands Key words Venous thromboembolism, treatment, composite outcome, LMW heparin, fixed-dose body weight-independent heparin regimen Summary Background. Body weight-adjusted subcutaneous low-molecularweight heparin (LMWH) has been proven to be at least as effective and safe as dose-adjusted intravenous unfractionated heparin (UFH) for the treatment of patients with venous thromboembolism. However, body weight-adjusted dosage of low-molecular-weight heparin may be cumbersome and could lead possibly to incorrect dosing. Therefore a fixed LMWH dose, independent of body-weight, might rationalize initial treatment for venous thromboembolism. Methods. Patients with proven proximal deep-vein thrombosis were randomly assigned to fixed dose subcutaneous LMWH Certoparin (8,000 anti-factor Xa U b.i.d.; 265 patients) or to adjusted dose i.v. UFH (273 patients) for 12 days. Vitamin K antagonists were started between day 3 and 7 and continued for up to 6 months. The primary outcome measure was a 30 percent or greater improvement in the Marder Score, as revealed by repeated venography on day 12 (end of the initial treatment). The secondary composite outcome measure included death, recurrent venous thromboembolism and major bleeding and was assessed at day 12 and after 6 months by a blinded adjunction committee. Results. The Marder score improved by 30% or more in 30.3% and 25.0% of patients assigned to LMWH (198 paired venograms) and UFH (192 paired venograms), respectively (2p = 0.26). At the end of the initial treatment, the composite outcome was observed in 4 of the 265 patients (1.5%) randomized to LMWH, as compared with 14 of the 273 patients (5.1%) randomized to UFH (2p = 0.03). At 6 months these figures were 6.8% and 12.8%, respectively (risk reduction 0.53, confidence interval , 2p = 0.02). Conclusion. Fixed dose subcutaneous LMWH certoparin is at least as efficacious as UFH in resolving proximal vein thrombosis. * The investigators and the centers participating are listed in the appendix, pp Correspondence to: Job Harenberg, M.D., Professor of Medicine, I. Dept. of Medicine, University Hospital Mannheim, Ruprecht-Karls University Heidelberg, Theodor-Kutzer-Ufer 1-3, D Mannheim, Germany Tel: +49/621/ , -2789; Fax: +49/621/ ; J-Harenberg@ t-online.de With respect to the composite outcome of death, recurrent venous thromboembolism and major bleeding LMWH treatment performed significantly better both during initial therapy and at a six-month follow-up. Introduction Anticoagulation with heparin in patients with venous thromboembolism is well established as the standard initial treatment. Unfractionated heparin (UFH) is administered intravenously or subcutaneously, but requires dose adjustments, usually on the basis of the activated partial thromboplastin time (1). Recently, low-molecular-weight heparins (LMWHs) have been shown to be equally effective and safe as UFH for both the treatment of deep-vein thrombosis (DVT) (2, 3) or pulmonary embolism (4, 5). LMWHs differ from standard UFH in that they have a higher ratio of anti-factor Xa to anti-factor IIa activity, greater bioavailability, longer plasma half-life, a more predictable anticoagulant response (6) and less adverse events such as heparin induced thrombocytopenia (7). Most studies with LMWHs have been performed using body weight adjusted dosages. However, it is unclear from a pharmacokinetic and clinical point of view, whether body weight adjustment of LMWHs is really necessary for the treatment of acute DVT. One dose-finding study has been conducted in these patients without resulting in a definite relation of dosage of LMWH and clinical outcome (8). In clinical practice confusion may arise with the use of different doses of LMWH and errors may occur in prescription, especially when patients are treated out of hospital by self-injection. For the LMWH certoparin, the impact on the Marder score in DVT patients was first shown in two pilot studies (8, 9). In a subsequent larger clinical trial certoparin given subcutaneously twice daily in a fixed therapeutic dose was associated with a more pronounced reduction of thrombus size as compared to aptt-controlled UFH. However, the effect of this dose regimen on long term clinical outcome was not investigated (10). So far, clinical studies comparing LMWH with UFH in the treatment of acute DVT documented the incidence of death, recurrent venous thromboembolism and major bleeding complications, separately (2, 3, 5). However, these outcomes are all relevant clinical events to patients. The purpose of this study was to assess both thrombus regression by repeated venography and to compare the LMWH certoparin with UFH regarding the composite outcome of death, recurrent venous thromboembolism and major bleeding. 652

2 Harenberg et al.: Fixed-dose SC LMWH Therapy Patients and Methods Study Design The study was a multicenter randomized clinical trial comparing fixed-dose, body weight-independent, subcutaneous LMWH certoparin with aptt-controlled intravenous UFH in patients with acute proximal deep-vein thrombosis. Thirty-three centers in Germany, Austria, Switzerland and the Czech Republic recruited patients in this trial. Acute deep-vein thrombosis is routinely treated with dose adjusted intravenous UFH in hospitalized patients in these countries for a period of 12 to 15 days. This study was carried out according to the guidelines of the Declaration of Helsinki. The study protocol was approved by all institutional review boards. All patients had to give written informed consent prior to randomization. Patients Hospitalized patients 30 years of age with acute symptomatic proximal deep-vein thrombosis (i.e. thrombosis in the popliteal vein or a more proximal vein) documented by ascending venography (11) were eligible. Patients were excluded from the study if they had one of the following: indication for surgical or fibrinolytic treatment of deep-vein-thrombosis, duration of symptoms for more than 3 weeks, ongoing oral anticoagulation, renal failure, severe hypertension (blood pressure >200 mmhg systolic and >105 mmhg diastolic while on anti-hypertensive treatment), severe hepatic failure, currently active bleeding or disorders contraindicating anticoagulant therapy, contraindication to oral anticoagulants, pregnancy, known intolerance to heparins, intolerance to contrast media, any operation within the past 8 days, acute severe pulmonary embolism, platelet count <100,000/ l, treatment with heparin >24 h before inclusion into the study and treatment with platelet-inhibiting drugs ( 100 mg acetylsalicylic acid daily allowed). Treatment Regimens Patients assigned to therapy with LMWH received a fixed dose of 8,000 international anti-factor Xa units (axa IU) of certoparin (Mono-Embolex NM) subcutaneously b.i.d. Patients assigned to therapy with UFH received an initial bolus of 5,000 IU, followed by continuous intravenous infusion at an initial rate of 20 IU/kg/h. The dose of UFH was subsequently adjusted to a target activated partial thromboplastin time (aptt) of two- to threefold the reference value. The aptt-tests were performed 4-6 hrs after the start of treatment and thereafter once daily; whenever a subtherapeutic coagulation time had been determined, the dose was adjusted and aptt was repeated within 6 h. The heparin treatment period lasted 12 days ranging from 7 to 15 days. In all participating countries it is common practice to give initial heparin treatment for a period of 12 to 15 days. This regimen is also based on the results from clinical studies having shown that a 5-day course of iv-heparin (12, 13) was as effective and safe as a 10-day period. Finally, the randomized design of the present study and the comparable duration of the treatment with iv heparin (10.7 ± 2.7 days) and LMWH (11.8 ± 2.8 days) add to the internal validity of the results. Oral anticoagulation (vitamin K antagonists) was started between the 3rd and 7th day and was continued for up to 6 months. Treatment with low-molecular-weight heparin or heparin was stopped as soon as an International Normalized Ratio (INR) of 2.0 or more was achieved for a minimum of 2 consecutive days. Surveillance and Follow-up All patients were examined daily during the heparin treatment period. Signs and symptoms of recurrent venous thromboembolism or hemorrhage were evaluated. Patients with suspected recurrent deep-vein thrombosis underwent ascending venography (11). Patients in whom pulmonary embolism was clinically suspected underwent ventilation-perfusion scanning. Pulmonary embolism was diagnosed if a high probability perfusion defect was documented (14). Blood counts were obtained twice weekly during the initial treatment period and whenever bleeding occurred. Severe thrombocytopenia was defined as a decrease of platelet count below 50% of the initial value or below 80,000/ l during the study. Bleeding was defined as major, if it was overt and associated either with decrease of the hemoglobin concentration by at least 2 g/dl or if a transfusion of 2 or more units of blood was indicated or if the bleeding was intracranial, retroperitoneal or intraspinal. All patients were contacted after 6 months. A check-list was used to evaluate data on signs and symptoms of recurrent venous thromboembolism, mortality, and hemorrhage. In case of death, autopsy was performed whenever permission was obtained. Outcome Measures All patients underwent a second ascending venography of the affected leg between days 7 and 15, unless clinical symptoms prompted an earlier test. Venographies were evaluated by two independent radiologists unaware of the treatment received and the order in which the tests were obtained. The thrombosed veins were assessed quantitatively according to the scoring system of Marder et al. (15). The primary outcome measure was defined as a relevant reduction of the Marder score of 30% on the second venography as compared to the one obtained at entry. The choice of this relevant reduction was also based on the outcome of earlier studies (8, 10, 16, 17), who used 20% or 30% reduction of thrombus size as primary outcome. The secondary outcome measure of the study was the composite outcome of death, recurrent venous thromboembolism, and major bleeding during the treatment period and after 6-month test follow-up. Data on all potential outcome events were evaluated by an independent adjunction committee, which was unaware of the treatment assignment. Statistical methods. The analysis was performed on an intention-to-treat basis. The results in the two treatment groups were compared with Fisher s exact t-test (age, weight) and the Wilcoxon test. Ninety five percent confidence intervals for the differences between the 2 groups and incidence of the outcome events were calculated with normal approximation to the binomial distribution. Based on earlier data (10), it was estimated that a sample containing 180 evaluable venographies per group would achieve a power of 80% in detecting a treatment difference for the improvement of the Marder Score of 30% on the second venogram compared to the venogram at entry. The log-rank test was used to assess differences in the cumulative incidences of the secondary outcomes. Results From May 1996 to June 1998, 541 patients were eligible for the study. Three patients withdrew informed consent before receiving any study medication. Of the remaining 538 patients, 265 were assigned to receive low-molecular-weight heparin and 273 received intravenous unfractionated heparin. No differences in the baseline characteristics of the two treatment groups could be detected (Table 1). Table 1 Baseline characteristics of study patients (mean values ± standard deviation or number of patients and percentage) * Patients with more than 1 event were counted only once. 653

3 0.26, relative risk [RR] 0.93, confidence interval [CI] , exact Fisher s test). The absolute values of the Marder score were 23.6 ± 8.3 at entry and 19.8 ± 9.9 after treatment with LMWH and 24.2 ± 8.2 and 21.4 ± 9.7 after therapy with UFH (2p = 0.18, Wilcoxon test). Composite Outcome Fig. 1 Kaplan-Meier curves showing the incidence of the composite outcome events during 6 months (2p = 0.02 by the log-rank test) A total of 529 of 538 patients (98.3%) completed the 6-month follow-up. 261 of 265 patients were originally assigned to LMWH and 268 of 273 patients to UFH. Anticoagulant Therapy The mean (± SD) duration of anticoagulant treatment was 11.8 ± 2.8 days with LMWH and 10.7 ± 2.7 days with UFH, respectively (not significant). The doses of UFH were 1352 ± 340 IU per h at day 1 and 1427 ± 476 IU per h at day 10. Oral anticoagulation was started at day 8.2 ± 2.4 in patients on LMWH and at day 7.9 ± 2.1 in patients on UFH. Primary Outcome Venographies at baseline and between days 7 to 15 were evaluable in 390 patients (198 LMWH and 192 UFH patients). The Marder score was reduced by 30 percent or more on the second venography in 30.3% of patients treated with LMWH and 25.0% treated with UFH (2p = At least one of the composite outcome measures of death, recurrent venous thromboembolism or major bleeding was observed during the heparin treatment period in 4 (1.5%) patients receiving LMWH and in 14 (5.1%) patients receiving UFH (Table 2; 2p = 0.03). From randomization to 6 months at least one event occurred in 18 patients (6.8%) allocated to LMWH and in 35 patients (12.8%) allocated to UFH (2p = 0.02, Table 2). The cumulative incidence of the composite outcome is shown in Fig. 1. In addition, the day of occurrence of the respective elements in the two treatment groups is given in Table 3. Other Findings The dose and the body weight as well as the changes of the Marder score did not relate to incidences of death, recurrent venous thromboembolism, major bleeding complications, and minor bleeding complications in patients allocated to LMWH and UFH during the initial treatment period (data not shown). Severe thrombocytopenia without thromboembolism developed in 3 patients receiving LMWH and in 4 patients receiving UFH. Discussion Previous clinical studies have indicated that LMWHs given subcutaneously are at least as effective and safe as intravenous aptt-controlled UFH for the treatment of patients with acute DVT (2, 8, 10, 18, 19). Patients were treated with once or twice daily subcutaneous injection of LMWH, the dose of which was adjusted to the patient s Table 2 Incidence of composite outcome events of death, recurrent VTE and major bleedings in the study patients according to the treatment group 654

4 Harenberg et al.: Fixed-dose SC LMWH Therapy Table 3 Causes and day of occurrence after start of therapy of the composite outcome of death, recurrent venous thromboembolism (VTE) and non-fatal major haemorrhage in the 2 treatment groups body weight. None of the studies suggested an increased bleeding risk up to dosages of 200 international anti-factor Xa units per kg per day. A single study reported an increased bleeding risk with higher doses of LMWH in patients with deep-vein thrombosis shortly after surgery (20). Therefore, the need for dose adjustments of the LMWH on the basis of body weight is not established and it is also not definitively supported from data of pharmacokinetic studies. Previous meta-analyses revealed a significant reduction in the incidence of thrombus extension in favour of LMWH as compared to doseadjusted intravenous heparin (18, 19). Also, mortality, recurrent thromboembolic events and major hemorrhages showed a non-significant trend in favor of LMWH. The clinical relevance of small improvements in the Marder score may be limited. We therefore decided, to use a substantial reduction of the thrombus size as the primary outcome of the study, represented by a 30% or greater reduction of the Marder score on the second venogram as compared to the venogram obtained at entry. Using this strict criterion we observed a non-significant trend to a greater responder rate in patients assigned to LMWH as compared to patients treated with UFH (30% vs. 25%). Limited information is available on the use of a fixed-dose, body weight independent, subcutaneous administration of LMWH in patients with acute DVT. Only one study (10) indicated that the fixed dose of 8,000 IU LMWH certoparin b.i.d. was effective in reducing venographic thrombus extension. Therefore, in the present study the same fixeddose body weight-independent subcutaneous therapy regimen was evaluated and in addition to changes of venography, patients were followed up for clinical outcome. In the present study, the occurrence of the composite outcome consisting of death, recurrent thromboembolism and major hemorrhage occurred significantly less in patients assigned to LMWH, compared to patients allocated to UFH both during the initial treatment phase, as well as after 6-month follow-up (Table 2). So far, clinical trials comparing LMWH with UFH in the treatment of acute DVT documented the incidence of death, recurrent venous thromboembolism and major bleeding complications separately with the exception of one trial (4). In our view, these outcomes are all major adverse events from a patients perspective and ought to be combined. Because the present trial was conducted in an open manner, particular care was taken to minimize the potential for bias. Patients were included at each clinical center using central telephone randomization and almost 99% follow-up for all randomized patients was achieved. Furthermore, all clinical outcome events had to be confirmed by objective tests. Finally, all events were evaluated by an independent adjudication committee. An important issue is the safety of the fixed subcutaneous dose of certoparin in underweight patients. One of seven patients with body weights below 55 kilogram allocated to certoparin and one of five patients, also below 55 kilogram of body weight randomized to UFH, experienced major bleeding complications. This data in underweight patients indicated that the risk of bleeding in patients allocated to UFH was as high as in patients randomized to LMWH, even though the UFH-treatment was body weight-adjusted and aptt-controlled. In patients with a body weight over 100 kilogram, no recurrent venous thromboembolism occurred in the present study using fixed-dose body weight-independent LMWH-treatment. It is concluded that in patients with acute proximal deep-vein thrombosis a fixed body weight independent dose of 8,000 IU certoparin b.i.d. is at least as efficacious as intravenous UFH with regard to regression of thrombus size. This LMWH regimen appears to be clinically superior to UFH with regard to the composite outcome of death, recurrent venous thromboembolism and major bleeding both during initial treatment and over a period of 6 months. This fixed-dose regimen with LMWH further simplifies treatment and may lead to increasing convenience both for patients and medical personal. Appendix Writing Committee J. Harenberg, M.D., Mannheim; H. R. Büller, M.D., Amsterdam; M. V. Huisman, M.D., Leiden 655

5 Steering Committee: J. Harenberg, M.D., Mannheim; H. K. Breddin, M.D., Frankfurt; J. W. ten Cate, Amsterdam; V. Cepelak, M.D., Pilsen; M. Gräve, M.Sc., Nuremberg; M. V. Huisman, M.D., Leiden; K. Koppenhagen, M.D., Berlin; A. Tolle, M.D., Nuremberg; G. Weidinger, Ph.D., Nuremberg Critical event and safety committee: S. Haas, M.D., Munich; K. Koppenhagen, M.D., Berlin; J. Harenberg, M.D., Mannheim Central reading committee of venographies: K. Koppenhagen, M.D., Berlin; K. Fobbe, M.D., Berlin Biostatistics: M. Gräve, M.Sc., Nuremberg; H. Bachmann, M.Sc., Nuremberg Sponsor: Novartis Pharma GmbH, Nuremberg The following investigators, all in Germany, except as otherwise noted, also participated in the study. Study centers: J. A. Schmidt, M.D., Ch. Kirchhof, M.D., Magdeburg; U. Hoffmann, M.D., K. Huck, M.D., G. Huhle, M.D., Ch. Krahl, M.D., J. Schmitt, M.D., J. Schmitz, M.D., B. Simonis, M.D., Mannheim; F. Patek, M.D., F. Kulic, M.D., Usti na Labem; K. Kuhn, M.D., G. Straeten, M.D, Stendal; G. Nissler, M.D., R. A. Reinicke, M. Irani, Hofheim; M. Vambera, M.D., L. Pesl, M.D., P. Jelinek, M.D., Ceske Budejovice; V. Cepelak, M.D., M. Novák, V. Benes, Pilsen; T. Höfs, M.D., K. Bensch, M.D., H.-J. Presser, M.D., Magdeburg; E. Dundalek, M.D., P, Spürk, M.D., Menden; W. Brockhaus, M.D., R. Sebiger, M.D., Nuremberg; K. L. Schulte, M.D., U. Sakriß, M.D., B. Bäsell, Berlin; W.-D. Pietruschka, M.D., Neubrandenburg; F. Liebold, M.D., L. Künzel, M.D., Leipzig; H. Darius, M.D., W. Ibe, M.D., Mainz; M. Kunze, M.D., H. Boden, M.D., A. Öhring, M.D., Suhl; K. Sondern, M.D., Dortmund; W. Mokros, M.D., K. Siedentopf, M.D., Magdeburg; C. Kelbel, M.D., F. Houchangnia, M.D., Lüdenscheid; T. Wagner, M.D., H. J. Siemens, M.D., Lübeck; B. Kohler, M.D., J. Fichtner, M.D., Bruchsal; R. Zahn, M.D., E. Fromm, M.D., Ludwigshafen; J. Stumpf, M.D., M. Schöler, M.D., Schesslitz; N. Zinnagel, M.D., K. Forstner, M.D., Salzburg; G. Cieslinski, M.D., J. Reuss, M.D., Frankfurt; K. Rhyner, M.D., Ch. Herpertz, M.D., Glarus; S. Semrad, M.D., P. Kala, M.D., R. Dvorák, M.D., Brno-Bohunice; H. Volkmann, M.D., G. Lauke, M.D., Annaberg-Buchholz; M. Penka, M.D., D. Záckova, M.D., Brno-Bohunice; Th. Scholten, M.D., S. Bußhoff, Hagen; R. Lorenz, M.D., München. References 1. Hyers TM, Agnelli G, Hull R, Weg JG, Morris TA, Sammama M, Tapson V. Antithrombotic therapy for venous thromboembolic disease. Chest 1998; 114: Levine M, Gent M, Hirsh J, Leclerc J, Anderson D, Weitz J, Ginsberg JS, Turpie AG, Demers C, Kovacs M. A comparison of low-molecular-weight heparin administered primarily at home with unfractionated heparin administered in the hospital for proximal deep-vein-thrombosis. N Engl J Med 1996; 334: Koopman MMW, Prandoni P, Piovella F, Ockford PA, Brandjes DP, Van der Meer J, Gallus AS, Simonneau G, Chesterman CH, Prins MH. Treatment of patients with proximal-vein thrombosis with intravenous unfrationated heparin in hospital compared with subcutaneous low-molecular-weight heparin out of hospital or with early discharge. N Engl J Med 1996; 334: Simonneau G, Sors H, Charbonnier B, Page Y, Laaban J-P, Azarian R, Laurent M, Hirsch J-L, Ferrari E, Bosson J-L, Mottier D, Beau B for the THÉSÉE Study Group. A comparison of low-molecular-weight heparin with unfractionated heparin for acute pulmonary embolism. N Engl J Med 1997; 337: The Columbus Investigators. Low-molecular-weight heparin in the treatment of patients with venous thromboembolism. N Engl J Med 1997; 337: Weitz JI. Low Molecular Weight Heparins. N Engl J Med 1997; 337: Ginsberg JS. Management of venous thromboembolism. N Engl J Med 1996; 335: Kirchmaier CM, Lindhoff-Last E, Rübsam D, Scharrer I, Vigh ZS, Mosch G, Wolf H, Breddin HK. Regression of deep vein thrombosis by iv.- administration of a low molecular weight heparin results of a pilot study. Thromb Res 1994; 73: Harenberg J, Huck K, Bratsch H, Stehle G, Dempfle CE, Mall K, Blauth M, Usadel KH, Heene DL. Therapeutic application of subcutaneous low-molecular-weight heparin in acute venous thrombosis. Haemostasis 1990; 20 (suppl. 1): Kirchmaier CM, Wolf H, Schäfer H, Ehlers B, Breddin HK for the certoparin-study group. Efficacy of a low molecular weight heparin administered intravenously or subcutaneously in comparison with intravenous unfractionated heparin in the treatment of deep venous thrombosis. Int Angiol 1998; 17: Rabinov K, Paulin S. Roentgen diagnosis of venous thrombosis in the leg. Arch Surg 1972; 104: Gallus A, Jackaman J, Tillett J, Mills W, Wycherley A. Safety and efficacy of warfarin started early after submassive venous thrombosis or pulmonary embolism. Lancet 1986, II: Hull RD, Raskob GE, Rosenbloom D, Panju AA, Brill Edwards P, Ginsberg JS, Hirsh J, Martin GJ, Green D. Heparin for 5 days as compared with 10 days in the initial treatment of proximal venous thrombosis. N Engl J Med 1990; 322; The PIOPED Investigators. Value of the ventilation/perfusion scan in acute pulmonary embolism: results of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED). JAMA 1990; 263: Marder VJ, Soulen RL, Atichartakan V, Budzynski AZ, Parulekar S, Kim IR. Quantitative venography assessment of deep vein thrombosis in the evaluation of streptokinase and heparin therapy. J Lab Clin Med 1977; 89: Prandoni P, Lensing AW, Buller HR, Carta M, Cogo A, Vigo M, Casara D, Ruol A, ten Cate JW. Comparison of subcutaneous low-molecular-weight heparin with intravenous standard heparin in proximal deep-vein thrombosis. Lancet 1992, 339; Brandjes DP, Heijboer H, Buller HR, de Rijk M, Jagt H, ten Cate JW. Acenocoumarol and heparin compared with acenocoumarol alone in the initial treatment of proximal-vein thrombosis. N Engl J Med 1992, 327; Lensing AWA, Prins MH, Davidson BL, Hirsh J. Treatment of deep venous thrombosis with low-molecular-weight heparins. A meta-analysis. Arch Intern Med 1995; 155: Siragusa S, Cosmi B, Piovella F, Hirsh J, Ginsberg JS: Low-molecularweight heparins and unfractionated heparin in the treatment of patients with acute venous thromboembolism: results of a meta-analysis. Am J Med 1996; 100: Bratt G, Törnbohm E, Graqvist St, Aberg W, Lockner D. A comparison between low molecular weight heparin (KABI 2165) and standard heparin in the intravenous treatment of deep venous thrombosis. Thromb Haemost 1985; 54: Received July 22, 1999 Accepted after resubmission January 17,