Critically Appraised Topic
Question: Are vena caval filters better for preventing pulmonary embolisms from a proximal DVT?
Appraiser: Eric W. Terman, MD
Date: 2/24/98
Clinical Bottom Lines:
1. Patients who had filters placed were less likely to have pulmonary embolisms, but they were more likely to have recurrent DVTs.
2. Even though patients, who did not get a filter placed, had pulmonary embolisms more often, they were not any more likely to die.
Methods: Multi-center, randomized, controlled trial using consecutive patients, were placed in either filter/no filter groups and then regular vs low molecular weight heparin groups. This resulted in four treatment groups.
Exclusion criteria included: prior filter, contra-indications for anticoagulation, an indication for thrombolysis, curative anticoagulation, allergy to iodine, short life expectancy, hereditary thrombophilia, severe renal or hepatic failure, pregnancy or noncompliance.
Coumadin therapy was initiated on the fourth day and continued for at least three months.
Patients were initially included by a positive angiogram or a high prob. VQ scan.
Results:
| Time period | symptomatic PE | recurrent DVT | Death | |
| Filter | 0-3 mos | 2 | 9 | 15 |
| 3 mos- 1 yr. | 0 | 8 | 12 | |
| 1-2 yrs | 4 | 20 | 16 | |
| All | 6 (3.4%) | 37 (20.8%) | 43 (21.6%) | |
| No filter | 0-3 mos | 6 | 6 | 10 |
| 3 mos- 1 yr. | 4 | 7 | 12 | |
| 1-2 yrs | 2 | 8 | 18 | |
| All | 12 (6.3%) | 21 (11.6%) | 40 (20.1%) | |
| p-value | 0.16 | 0.02 | 0.65 | |
| NNT or NN to harm | 34, but not truly sig. so difficult to calculate | 11 treated to harm 1 | No difference | |
Comments:
1. The study was well constructed, but was not blinded. Nevertheless hard end points such as death, major bleeding, and recurrent DVT were followed, making it less subject to bias.
2. The study population was French, so it might be more homogeneous than the United States’ population.
3. There were some cross overs, i.e. 8 patients assigned to no filter, did get one within 12 days secondary to having pulmonary embolisms. The study was analyzed on an intention to treat basis. This is good, except that the eventual mortality in the filter group may have been altered. This may have been balanced by the fact that four of the patients assigned to the filter group then refused it.
4. Overall there does not seem to be a benefit in placing a filter in the average person who presents with a proximal DVT.
Source:
Decousus H, Leizorovicz A, Parent F, Page Y, Tardy B, Girard P, Laporte S, Faivre R, Charbonnier B, Barral FG, Huet Y, Simonneau et al. A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis, NEJM. 338(7):409-415.