Search tips
Search criteria 


Logo of bmjThis ArticleThe BMJ
BMJ. 2007 March 31; 334(7595): 645.
PMCID: PMC1839231

Anticoagulation for venous thromboembolism

John W Eikelboom, thrombosis physician, Jeffrey S Ginsberg, thrombosis physician, and Jack Hirsh, thrombosis physician

Longer duration of treatment does not reduce risk of recurrence unless continued indefinitely

The optimal duration of anticoagulant therapy for the treatment of venous thromboembolism has been the subject of many randomised trials over the past 15 years.1 2 The results indicate no clear advantage for many patients of prolonging warfarin beyond three to six months, because of the risk of bleeding and the inconvenience. The annual incidence of major bleeding in patients who take warfarin for longer than three months is 2-3%, with an estimated case fatality rate of 9.1% (95% confidence interval 2.5% to 21.7%).3 Also, duration of treatment has little effect on the long term risk (after the first three months) of recurrence. Trials have shown that the frequency of recurrence at two to three years is similar in patients taking three months or 12 months of oral anticoagulation.1 2 Whether the frequency of recurrence would be the same after longer lengths of treatment and follow-up is not known. Consequently, the optimal duration of oral anticoagulant treatment remains a contentious issue.

In this week's BMJ, a randomised controlled trial by Campbell and colleagues4 investigates the optimal duration of oral anticoagulant treatment. It compared three or six months of warfarin (target international normalised ratio 2.0-3.5) after an initial five days of heparin in 749 patients with suspected or confirmed venous thromboembolism without ongoing risk factors for recurrence. After 12 months, recurrent fatal or non-fatal venous thromboembolism occurred in 8% of patients in each treatment group (difference 0%, −3.1% to 4.7%, P=0.80). Major bleeding occurred in significantly more patients taking warfarin for six months than for three months (2% v 0%; difference 2%, 0.7% to 3.5%, P=0.008). The trial was discontinued prematurely because of slow recruitment, but the results provide no evidence of benefit and clear evidence of harm (bleeding) of longer duration of treatment. These results are consistent with other studies and a meta-analysis of individual patient data,5 which found similar frequencies of recurrence after discontinuing warfarin in patients given at least three months of anticoagulants.

The meta-analysis of individual patient data from five randomised trials compared different durations of anticoagulant treatment for venous thromboembolism, and confirmed the results of randomised trials that continuing treatment beyond three to six months does not reduce the risk of recurrence after warfarin is stopped.5 Each trial consistently showed a cluster of recurrences immediately after stopping treatment.5 The reasons for this are unknown, but possible explanations include hypercoagulability of the blood as a result of stopping warfarin6 or a continuing thrombogenic state in some patients.

On the basis of current evidence how should we treat our patients? Patients with a first episode of venous thromboembolism should receive warfarin for at least three months. The exception is patients with isolated distal vein thrombosis, in whom six weeks is generally adequate. Although long term treatment is highly effective for preventing recurrence in patients with unprovoked venous thromboembolism, a “catch-up phenomenon” occurs after warfarin is stopped, suggesting that long term warfarin does not alter the natural history of the disease. Because of this catch-up phenomenon, there is little point continuing treatment beyond three to six months, unless a continuing reversible risk factor exists, in which case treatment is continued until the risk is no longer present or a decision is made to continue treatment indefinitely. A decision to treat a patient indefinitely is reasonable in patients with a very high risk of recurrence, such as those with more than one episode of unprovoked thrombosis, those with cancer and thrombosis, and those with high risk thrombophilia. Indefinite treatment might also be considered in patients with severe post thrombotic syndrome and in those with a strong preference for minimising their risk of recurrence by continuing anticoagulants.

Because it is thought that long term warfarin yields a net benefit for patients at highest risk of recurrence and that stopping warfarin is reasonable in those with a low risk of recurrence, efforts have been directed towards identifying clinical and laboratory markers that better predict the recurrence risk.7 Ultimately, however, the question of which patients should be treated with anticoagulants indefinitely will require large randomised studies that have sufficient power to show a worthwhile reduction of morbidity or mortality, or an improvement in quality of life.


Competing interests: None declared.

Provenance and peer review: Commissioned; not externally peer reviewed.


1. Agnelli G, Prandoni P, Santamaria MG, Bagatella P, Iorio A, Bazzan M, et al. 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;345:165-9. [PubMed]
2. Agnelli G, Prandoni P, Becattini C, Silingardi M, Taliani MR, Miccio M, et al. Extended oral anticoagulant therapy after a first episode of pulmonary embolism. Ann Intern Med 2003;139:19-25. [PubMed]
3. Linkins LA, Choi PT, Douketis JD. Clinical impact of bleeding in patients taking oral anticoagulant therapy for venous thromboembolism: a meta-analysis. Ann Intern Med 2003;139:893-900. [PubMed]
4. Campbell IA, Bentley DP, Prescott RJ, Routledge PA, Shetty HG, Williamson IJ. Anticoagulation for three versus six months in patients with deep vein thrombosis or pulmonary embolism, or both: randomised trial. BMJ 2007. 10.1136/bmj.39098.583356.55
5. Boutitie F, Pinede L, Schulman S, Agnelli G, Raskob G, Julian J, et al. Incidence of recurrent venous thromboembolism according to duration of vitamin K antagonism: Results of a meta-analysis of individual patient data [abstract]. Blood 2004;104:a710
6. Palareti G, Legnani C. Warfarin withdrawal. Pharmacokinetic-pharmacodynamic considerations. Clin Pharmacokinet 1996;30:300-13. [PubMed]
7. Schulman S, Ogren M. New concepts in optimal management of anticoagulant therapy for extended treatment of venous thromboembolism. Thromb Haemost 2006;96:258-66. [PubMed]

Articles from The BMJ are provided here courtesy of BMJ Publishing Group