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


Vena cava filters reduced pulmonary embolism in high-risk patients

ACP J Club. 1998 Jul-Aug;129:6. doi:10.7326/ACPJC-1998-129-1-006

Source Citation

Decousus H, Leizorovicz A, Parent F, et al., for the Prévention du Risque d'Embolie Pulmonaire par Interruption Cave Study Group. A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis. N Engl J Med. 1998 Feb 12;338:409-15.



In patients with deep venous thrombosis (DVT), what is the effect of vena cava filters (vs no filter) and subcutaneous low-molecular-weight heparin (LMWH) (vs intravenous unfractionated heparin [UH]) on pulmonary embolism (PE) and recurrent DVT?


Randomized controlled trial (RCT) (factorial design).


44 hospitals in France.


400 patients ≥ 18 years of age (mean age 73 y, 52% women) who had confirmed proximal DVT and were at high risk for PE. Exclusion criteria were previous filters, failure of or contraindications to anticoagulants, receipt of anticoagulants for > 48 hours, need for thrombolysis, short life expectancy, allergy to iodine, hereditary thrombophilia, severe renal or hepatic failure, or pregnancy. Follow-up was ≥ 98%.


Patients were allocated to vena cava filters (n = 200) or no filter (n = 200) and to LMWH (enoxaparin) subcutaneously, 1 mg/kg of body weight daily (n = 195), or UH, intravenous bolus of 5000 IU followed by a continuous infusion of 500 IU/kg of body weight daily with adjusted dosing (n = 205) for 8 to 12 days. Beginning at day 4 and for ≥ 3 months, patients received graded compression stockings and warfarin or acenocoumarol or, if necessary, UH.

Main outcome measures

PE, recurrent DVT, death, and hemorrhage at 12 days and 2 years.

Main results

Intention-to-treat analysis was done. 4 patients assigned to a filter did not receive it; 8 patients in the no-filter group received a filter within 12 days. Patients in the filter group had a lower rate of PE at 12 days (P = 0.03) and a higher rate of DVT at 2 years (P = 0.02) than did patients in the no-filter group (Table). The groups did not differ at 12 days for major hemorrhage or death or at 2 years for PE, major hemorrhage, or death. For heparin, the groups did not differ for any outcome at 12 days or 2 years.


In patients with confirmed deep venous thrombosis, a vena cava filter reduced the incidence of pulmonary embolism at 12 days, but the incidence of deep venous thrombosis increased at 2 years. Low-molecular-weight heparin and unfractionated heparin did not differ for any outcome.

Sources of funding: Bellon Rhône-Poulenc Rorer Laboratories; Ministère Français de la Santé; Caisse Nationale d'Assurance Maladie; Structure Régionale d'Evaluation Interhospitaliére Rhône-Alpes, Fondation de l'Avenir; Laboratoire B. Braun France; Laboratories Diagnostica Stago France.

For correspondence: Dr. H. Decousus, Thrombosis Research Group, Clinical Pharmacology Unit, Hôpital Bellevue, CHU Saint-Etienne, 42055 Saint-Etienne CEDEX 2, France.

Table. Vena cava filters vs no filter to prevent pulmonary embolism (PE) for patients with proximal deep venous thrombosis (DVT)*

Outcomes Filters No filter RRR (95% CI) NNT (CI)
PE at 12 d 1.1% 4.8% 77% (8 to 94) 27 (13 to 323)
RRI (95% CI) NNH (CI)
DVT at 2 y 20.8% 11.6% 80% (11 to 195) 11 (6 to 63)

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


Although vena cava filters have been used for 3 decades, this study by Decousus and colleagues is the first RCT on use of these filters for the prevention of PE in patients with proximal DVT. It was meticulously done, follow-up was excellent, and it confirmed previous observations that LMWH is as effective and safe as UH for treating DVT or PE (1-3).

Patients allocated to UH and no filter had a 5% rate of PE at 12 days, which is consistent with the recent literature (1-3). Patients in the heparin-plus-filter group were one fifth as likely to have a PE at day 12, but the filter doubled the chance of having recurrent DVT at 2 years. The mortality rate in the 2 groups did not differ at 12 days or 2 years.

This study shows that filters do not prevent all PE. 37 patients with filters had symptomatic recurrent DVT or PE, and 16 had thrombosis at the filter site (12 had DVT, and 4 had PE). These results provide evidence for the need for anticoagulation therapy with filter placement.

The increased rate of recurrent DVT and the risk for further immobilization from the post-thrombotic syndrome suggest that physicians should consider selective use of filters. To prevent 1 additional early PE with a filter, 27 patients would need to be treated; however, with filter treatment, 2 to 3 additional cases of recurrent DVT would be expected. Data from RCTs that include quality-of-life and cost outcomes and that focus on these high-risk groups (4) would provide further guidance for clinicians who treat patients with DVT.

Wesley Ely, MD, MPH
Vanderbilt University Medical CenterNashville, Tennessee, USA


1. The Columbus Investigators. Low-molecular-weight heparin in the treatment of patients with venous thromboembolism. N Engl J Med. 1997;337:657-62.

2. Simonneau G, Sors H, Charbonnier B, et al. A comparison of low-molecular-weight heparin with unfractionated heparin for acute pulmonary embolism. The THESEE Study Group. Tinzaparine ou Heparine Standard: Evaluations dans l'Embolie Pulmonaire. N Engl J Med. 1997;337:663-9.

3. Weitz JI. Low-molecular-weight heparins. N Engl J Med. 1997;337:688-98.

4. Pacouret G, Alison D, Pottier JM, Bertrand P, Charbonnier B. Free-floating thrombus and embolic risk in patients with angiographically confirmed proximal deep venous thrombosis. A prospective study. Arch Intern Med. 1997;157:305-8.