MEDICAL-ON-LINE/ALAMY
Deep vein thrombosis (DVT) in the calf of a patient.
The Evidence
The 2008 ACCP guidelines on VTE management follow a grading system that classifies recommendations as Grade 1 (strong) or Grade 2 (weak), and classifies the quality of evidence as A (high), B (moderate), or C (low).12 The ACCP guidelines’ recommended first-line treatment for a confirmed DVT is anticoagulation with subcutaneous low-molecular-weight heparin, intravenous unfractionated heparin, monitored subcutaneous heparin, fixed-dose subcutaneous unfractionated heparin, or subcutaneous fondaparinux (all Grade 1A recommendations). The ACCP recommends against the routine use of an IVC filter in addition to anticoagulants (Grade 1A). However, for patients with acute proximal DVT, if anticoagulant therapy is not possible because of the risk of bleeding, IVC filter placement is recommended (Grade 1C). If a patient requires an IVC filter for treatment of an acute DVT as an alternative to anticoagulation, it is recommended to start anticoagulant therapy once the risk of bleeding resolves (Grade 1C).4
click for large version
The drawings above show the path of emboli from the lower extremities to the lung (left); Greenfield Filter placement in relation to the heart and lungs (above right); and emboli trapped in a Greenfield Filter.
The 2008 ACCP guidelines for IVC filter use have a few important changes from the 2004 version. First, the IVC filter placement recommendation for patients with contraindications to anticoagulation was strengthened from Grade 2C to Grade 1C. Second, the 2008 guidelines omitted the early recommendation of IVC filter use for recurrent VTE, despite adequate anticoagulation (Grade 2C).13
Only one randomized study has evaluated the efficacy of IVC filters. All other studies of IVC filters are retrospective or prospective case series.
The PREPIC study randomized 400 patients with proximal DVT considered to be at high risk for PE to receive either an IVC filter or no IVC filter. Additionally, patients were randomized to receive enoxaparin or unfractionated heparin as a bridge to warfarin therapy, which was continued for at least three months. The primary endpoints were recurrent DVT, PE, major bleeding, or death. The patients were followed up at day 12, two years, and then annually up to eight years following randomization.14 At day 12, there were fewer PEs in the group that received filters (OR 0.22, 95% CI, 0.05-0.90). However, at year two, there was no significant difference in PE development in the filter group compared with the no-filter group (OR 0.50, 95% CI, 0.19-1.33).
click for large version
The drawings above show the path of emboli from the lower extremities to the lung (left); Greenfield Filter placement in relation to the heart and lungs (above right); and emboli trapped in a Greenfield Filter.
Additionally, at year two, the filter group was more likely to develop recurrent DVT (OR 1.87, 95% CI, 1.10-3.20). At year eight, there was a significant reduction in the number of PEs in the filter group versus the no-filter group (6.2% vs.15.1%, P=0.008). However, at eight-year followup, IVC filter use was associated with increased DVT (35.7% vs. 27.5%, P=0.042). There was no difference in mortality between the two groups.
In summary, the use of IVC filters was associated with decreased incidence of PE at eight years, offset by higher rates of recurrent DVT and no overall mortality benefit.14,15 Importantly, the indications for IVC filter use in this study differ from the current ACCP guidelines; all patients were given concomitant anticoagulation for at least three months, which might not be possible in patients for whom the ACCP recommends IVC filters.
There are no randomized studies to compare the efficacy of permanent IVC filters and retrievable filters for PE prevention. A retrospective study comparing the clinical effectiveness of the two filter types reported no difference in the rates of symptomatic PE (permanent filter 4% vs. retrievable filter 4.7%, P=0.67) or DVT (11.3% vs. 12.6%, P=0.59). In addition, the frequency of symptomatic IVC thrombosis was similar (1.1% vs. 0.5%, p=0.39).16 A paper reviewing the efficacy of IVC filters reported that permanent filters were associated with a 0%-6.2% rate of PE versus a 0%-1.9% rate with retrievable filters.7 Notably, these studies were not randomized controlled trials—rather, case series—and the indications for IVC filters were not necessarily those currently recommended by the ACCP.
