Current issues of ACP Journal Club are published in Annals of Internal Medicine


3 months of anticoagulation was more effective than 4 weeks for reducing recurrence of thromboembolism

ACP J Club. 1993 May-June;118:78. doi:10.7326/ACPJC-1993-118-3-078

Source Citation

Research Committee of the British Thoracic Society. Optimum duration of anticoagulation for deep-vein thrombosis and pulmonary embolism. Lancet. 1992 Oct 10; 340:873-6.



To compare the effectiveness of 4 weeks of anticoagulation with 3 months of anticoagulation in hospitalized patients with acute deep-vein thrombosis, pulmonary embolism, or both.


Randomized controlled trial of 12 months duration.


Multicenter study in the United Kingdom.


712 hospitalized patients (mean age 58 y, 53% men) with symptoms and signs suggesting deep-vein thrombosis, pulmonary embolism, or both and for whom heparin and warfarin treatment were indicated. Exclusion criteria were use of streptokinase, pulmonary embolectomy, neoplasia, polycythemia, thrombocytosis, deep-vein thrombosis or pulmonary embolism in the past 3 years, pregnancy, any disorder resulting in long-term immobility or confinement to bed, or need for long-term anticoagulation. 92% completed 12 months follow-up.


After initiation of heparin therapy, 358 patients were treated with warfarin for 4 weeks and 354 were treated with warfarin for 3 months. Heparin was usually given for 5 to 7 days and warfarin therapy was started on day 3.

Main outcome measures

Resolution of thromboembolism, recurrence, and adverse effects.

Main results

No difference existed between the 3-month group and the 4-week group for failure to resolve thromboembolism (P = 0.10) (Table). More patients in the 4-week group had a recurrence than did patients in the 3-month group (P = 0.04) (Table). Deep-vein thrombosis or pulmonary embolism developed postoperatively in 116 patients. Among these patients, thromboembolism did not resolve in 2 patients (1 in each treatment group), and 1 patient in the 4-week group had a recurrence. By contrast, among medical patients (298 in each group), thromboembolism did not resolve in 12 patients (4%) in the 3-month group compared with 23 patients (8%) in the 4-week group, and patients in the 4-week group had more recurrences (27 [9%] vs 14 [5%], P < 0.002). 54 patients died. 7 deaths were caused by pulmonary embolism (3 patients in the 4-week group and 4 patients in the 3-month group). 1 patient in the 4-week group died of hemorrhage. The rate of major hemorrhage was 6 per 1000 patient-months.


In patients where deep-vein thrombosis or pulmonary embolism or both developed postoperatively, 4 weeks of anticoagulation appeared to be adequate. In medical patients anticoagulation for 3 months was more effective than anticoagulation for 4 weeks in reducing the number of thromboembolisms that failed to resolve and the number of recurrences.

Source of funding: Morriston Davies Trust.

For article reprint: Dr. I.A. Campbell, Sully Hospital, Hayes Road, Sully, South Glamorgan CF64 5YA, Wales, UK. FAX 222-530-881.

Table. 3-month vs 4-week anticoagulation for thromboembolism*

Outcomes at 1 y 3-month anticoagulation 4-week anticoagulation RRR (95% CI) NNT (CI)
Failure to resolve thromboembolism 3.7% 6.7% 45% (-4.5 to 71) Not significant
Recurrence of thromboembolsim 4.0% 7.8% 51% (8.7 to 73) 25 (13 to 179)

*Abbreviations defined in Glossary; RRR, NNT, and CI calculated from data in article.


The optimal duration of anticoagulant therapy of venous thromboembolic disease is unknown. Based on the risk for recurrent thrombotic episodes and the risk for hemorrhagic complications caused by the anticoagulant therapy, most clinicians recommend between 1 and 6 months of therapy.

In this study reductions were reported in both nonresolution and recurrent thromboembolic events in patients who received a 3-month course of therapy compared with 4 weeks of therapy. Objective documentation of the recurrent (as opposed to initial) thrombotic event, however, was sought in only 38% of patients who received warfarin for 4 weeks and 48% of patients who received warfarin for 3 months. Clinical diagnosis is unreliable in venous thromboembolic disorders, especially for recurrent disease (1-3), and could have overestimated the number of events and resulted in bias against any of the treatment regimens. When the results were restricted to the 21 patients with objective documentation of failure to resolve or recurrence, the 2 treatment durations did not differ significantly (3.9% vs 2%).

The investigators also reported on a post-hoc analysis of the rates of nonresolution and recurrence of thromboembolism in patients who developed their initial thromboembolism in the postoperative period and reported no difference between the 2 treatment groups. The small number of patients in this subset, however, precludes any recommendation about therapy duration.

In conclusion, the evidence from previous studies (4) and from this study suggests that patients should receive a minimum of 3 months of warfarin after an episode of venous thromboembolism. Based on this study and previous studies (5), the incidence of major hemorrhage is estimated at between 2% and 4%.

Moira Cruickshank, MD
University HospitalLondon, Ontario, Canada


1. Hull RD, Carter CJ, Jay RM, et al. The diagnosis of acute, recurrent, deep-vein thrombosis: a diagnostic challenge. Circulation. 1983;67:901-6.

2. Huisman MV, Buller H, ten Cate JW. Utility of impedance plethysmography in the diagnosis of recurrent deep-vein thrombosis. Arch Intern Med. 1988;148:681-3.

3. Leclerc JR, Jay RM, Hull RD, Hirsh J. Recurrent leg symptoms following deep vein thrombosis. A diagnostic challenge. Arch Intern Med. 1985;145:1867-9.

4. Hyers TM, Hull RD, Weg JG. Antithrombotic therapy for venous thromboembolic disease. Chest. 1992;102(Suppl):408S-25S.

5. Levine MN, Hirsh J, Landefeld S, Raskob G. Hemorrhagic complications of anticoagulant treatment. Chest. 1992;102(Suppl):352S-63S.