Meta-analysis: IVC filters should be considered for certain patients at high risk of PE


In a recent meta-analysis, Yang Liu, Huan Lu (Henan Cancer Hospital, Zhengzhou, China), and colleagues found insufficient evidence to prove that inferior vena cava (IVC) filters can reduce pulmonary embolism (PE)-related mortality and overall mortality. However, they did find that IVC filters decrease the occurrence of PE without increasing deep venous thromboembolism (DVT) and major bleeding. Therefore, they conclude: “IVC filters should be considered after balancing the benefits and risks for the patients with contraindications to anticoagulant therapy of high risk for PE.”

Writing in the Journal of Vascular Surgery: Venous and Lymphatic Disorders, Liu, Lu, et al outline the scale of the issue at the core of their analysis: “Venous thromboembolism (VTE), which mainly presents as DVT and PE, is an important and potentially fatal disease with a high incidence and causes a huge burden.”

Considering treatment options, they write that, besides anticoagulation, the application of IVC filters is “commonly used” in patients with contraindications to anticoagulation therapy—for example, those who have been injured or who are at high risk of bleeding—or patients with a high risk of embolism occurrence. “However,” they write, “there is no clear consensus on the benefits and risks of IVCs from randomised controlled trials (RCTs)”.

Liu, Lu, and colleagues elaborate: “Systematic reviews have shown that IVC filters appear to reduce the incidence of subsequent PE, while appearing to increase the chance of DVT, and they have a limited effect on overall mortality. However, data in half of the included trials were not pooled into the effect size (odds ratio), which weakens the evidence of the conclusion.” Due to this lack of clarity on the subject, the researchers performed a systematic review and meta-analysis of RCTs with “more appropriate effect measures and more trials” in order to investigate the effect of IVC filters on PE-related mortality and complications.

The team searched PubMed and Cochrane libraries from inception to 31 October 2019 to identify RCTs for their meta-analysis. They detail that the primary outcome was mortality related to PE; secondary outcomes were overall mortality, occurrence of PE, DVT, and major bleeding.

Overall, the investigators included seven articles, comprising 1,274 patients, in their meta-analysis. They report that there was no significant difference in mortality related to PE between the IVC filter groups and the control group within three months (risk difference, -0.01; 95% confidence interval [CI], -0.03–0; p=0.11) and during the whole follow-up time with low heterogeneity (I2=0%).

However, they relay that the rates of new occurrence of PE within three months and during the whole follow-up period were lower in the IVC filter group than those in the control group (0.81% vs. 5.98%; risk ratio, 0.17; 95% CI, 0.04–0.65; p=0.01; 3.2% vs. 7.79%; risk ratio, 0.42; 95% CI, 0.25–0.71; p=0.0001, respectively).

Finally, Liu, Lu, et al communicate that there were no significant differences in the rates of new occurrence of DVT, major bleeding, and mortality rates during the whole follow-up period between the groups (p>0.05).

The authors acknowledge some limitations of the present study. They recognise, for example, that patients who were recruited in the included trials were “not completely consistent” regarding disease conditions and that performing subgroup analyses was limited because of the paucity of trials. Finally, they state that performance bias was a high risk, which was attributed to the “dramatic difference” in therapy between the two groups. “This made participants and personnel blinding from the intervention strategy impossible, and might have affected the outcomes to a certain extent,” they remark.

Looking forward, Liu, Lu, and colleagues suggest that large RCTs are required “to provide more robust evidence”.


Please enter your comment!
Please enter your name here