The Journal of Informed Pharmacotherapy 2001;7:200-203.
Download a PDF version of this article
Reviewers_Name: Cynthia A. Jackevicius, Jeff Nagge
Reviewers_Email: Cynthia.Jackevicius@uhn.on.ca; Jeff.Nagge@uhn.on.ca
Agnelli G, Prandoni P, Santamaria MG, Bagatella P, Iorio A, Bazzan M, Guazzaloca G, Bertoldi A, Tomasi A, Scannapieco G, Ageno W. Three months versus one year of oral anticoagulant therapy for idiopathic deep venous thrombosis. Warfarin Optimal Duration Italian Trial Investigators. N Engl J Med. 2001 Jul 19;345(3):165-9. PubMed Citation
This open-label, multicenter trial was conducted to determine the clinical
benefit of extending the three-month course of oral anticoagulant therapy for a
first episode of idiopathic proximal deep vein thrombosis (DVT) to
a treatment duration of one year. Patients between the ages of 18
and 85 were consecutively enrolled into
the study after a first episode of symptomatic, idiopathic proximal
DVT. After completing three months of uninterrupted therapy with warfarin
or acenocoumarol (target INR 2.0-3.0), eligible patients were randomized to
either discontinue oral anticoagulant therapy, or to continue it for an
additional nine months. The primary outcome for
this study was the symptomatic, objectively confirmed recurrence of
venous thromboembolism during a minimum
of two years of follow-up.
1. Was assignment of patients randomized?
Yes. Patients were randomized after three uninterrupted months of oral anticoagulant therapy without having a recurrence of thromboembolism or bleeding episode.
Yes. Two years of follow-up data for all 267 patients enrolled in the study was available. In addition, the primary analysis was performed on an intention-to-treat basis.
1. How large was the treatment effect?
Recruitment was terminated prematurely when the results of a planned interim analysis indicated the treatment effect did not meet the a priori criterion for effectiveness. The intention-to-treat analysis revealed that 21 of the 134 patients assigned to 12 months of therapy experienced an episode of RVTE (15.7%, average follow-up, 37.8 months), as did 21 of the 133 patients who were assigned to the short course therapy treatment arm (15.8%, average follow-up, 37.2 months). The calculated relative risk was 0.99 (95% confidence interval (CI), 0.57 to 1.73). Per-protocol analysis yielded similar results: 18 of 115 (15.7%) patients assigned to 12 months of therapy , and 21 of 126 (16.7%) assigned to 3 months of therapy experienced RVTE (relative risk 0.94 (95% CI 0.54 to 1.67)). The average time from randomization to RVTE was 11.2 months for those patients assigned to the 3-month treatment arm and 16.0 months in those assigned to 12-month treatment arm.
Major bleeding was defined as: clinically overt and associated with either a decrease in the hemoglobin level of at least 2g/dL or the need for transfusion of >=2 units of red cells, if the bleeding episode was retroperitoneal or intracranial in nature, or if the episode warranted permanent discontinuation of the study drug. Four patients (3.0%) assigned to 12-month therapy experienced nonfatal episodes of major bleeding during oral anticoagulant therapy. None of these patients had an INR above 3.0 at the time of bleeding. Two patients (1.5%) assigned to 3-month therapy suffered fatal bleeding. One episode was an intracranial hemorrhage one month after discontinuation of oral anticoagulation, and the other was a gastrointestinal bleed that occurred 12 months after discontinuation.
2. How precise was the estimate of the treatment effect?
The confidence intervals are quite wide (95% CI 0.57 to 1.73 for the intention-to-treat analysis). This is probably reflective of the relatively small sample size. There does not appear to be a significant skew in the CI that would suggest a possible benefit in the treatment group.
2. Were all clinically important outcomes considered?
Yes. Recurrent deep venous thromboembolism, major bleeding, and death were assessed. The location of the recurrent DVT events was not specified in the article. Venous thrombosis of the proximal veins of the legs carries a higher risk of developing pulmonary embolism than thrombosis of the distal veins. (2) Personal communication with the authors reveals that all 16 RVTE that occurred in those patients receiving 12 months of treatment were proximal in nature. In the 3 month treatment group, 16 of 18 (89%) RVTE were proximal, including popliteal vein location while the remaining events were distally located. (3) Exclusion of the two distal RVTE makes it even less likely that a potential treatment effect was missed.
The optimal duration of therapy with oral anticoagulants for patients with idiopathic venous thromboembolism has yet to be determined. Results from earlier studies have identified that this population is at increased risk of recurrent thromboembolism. (5,6) The Sixth ACCP Consensus Conference on Antithrombotic Therapy recommends that “patients with a first episode of idiopathic RVTE should be treated for at least 6 months” (7), but it is not clear whether a longer duration of treatment would be advantageous for this high risk group.
Recently, Kearon et al (1) compared an additional 24 months of warfarin therapy with placebo in patients with a first episode of idiopathic thromboembolism who had completed 3 months of initial anticoagulation. They demonstrated a 95 % reduction in the risk of recurrent thromboembolism in patients assigned to extended treatment with warfarin, but did not address whether this advantage would be maintained upon discontinuation of therapy. The current trial was designed to determine if the benefits of extended anticoagulation persists after therapy is discontinued. This trial supports the findings of Kearon et al (1) in that RVTE rates were reduced while patients received continued anticoagulation therapy. However, the results were not maintained after discontinuation, suggesting that prolonging the course of anticoagulation simply delays recurrence until treatment is discontinued. The focus of future trials may need to shift from the determination of an optimal duration of therapy for all patients to the identification of those patients at increased risk of RVTE who would benefit from indefinite anticoagulation. Patients at highest risk of recurrence must be identified due to the risks, costs and inconvenience associated with long-term anticoagulation therapy.
Copyright © 2001 by the Journal of Informed Pharmacotherapy. All rights reserved.