Vena Cava Filters Appear to Benefit Only a Small Fraction of Patients With Acute VTE
Use of inferior vena cava filters is associated with a lower risk of short-term mortality only in the small subgroup of patients hospitalized with acute venous thromboembolism (VTE) who have a temporary contraindication to anticoagulation due to active bleeding, an observational analysis shows. That comes at the cost of a higher rate of recurrent deep vein thrombosis (DVT) in the first year, however.
In the much larger group of patients who are eligible for anticoagulation, use of the filters does not impact the risk of death but—as in patients with active bleeding—seems to increase the risk of recurrent DVT, Richard White, MD (University of California Davis School of Medicine, Sacramento, CA), and colleagues report online ahead of print in Circulation.
The findings are consistent with the PREPIC 2 trial, which showed that filter use did not affect mortality or recurrent pulmonary embolism (PE) risk in patients with PE and no contraindication to anticoagulation, as well as with recent guidelines from the American College of Chest Physicians calling for the use of vena cava filters only in patients who cannot receive anticoagulant therapy, the authors say.
“In light of the American Board of Internal Medicine’s ‘Choosing Wisely’ initiative, the implications are straightforward: do not use a [vena cava filter] if your patient can receive standard anticoagulation therapy,” they write.
Although vena cava filters are frequently used for patients hospitalized for acute VTE, there is a lack of strong evidence supporting their use either in preventing death or lowering the risk of recurrent PE. In fact, some observational data show that using filters is associated with greater thrombotic and embolic risks.
To explore the utility of vena cava filters, White and colleagues performed a retrospective analysis of data on 85,159 patients without cancer who were admitted to nonfederal California hospitals for acute VTE between 2005 and 2010. That group included 80,697 patients with no contraindications to anticoagulation and 3,017 and 1,445 with temporary contraindications due to active bleeding or major surgery shortly before or during the index hospitalization, respectively. Rates of vena cava filter use were 9.6%, 36.2%, and 33.8% across groups. No information was available on the use, intensity, or duration of anticoagulation therapy.
After adjustment for propensity score using inverse probability weighting, as well as immortal time bias (which stems from variations in when the filters were inserted during the index hospitalization), use of filters was not associated with risk of 30-day mortality in patients eligible for anticoagulation or in those who underwent major surgery. Mortality at 90 days was higher among patients who received a filter and were able to take anticoagulation (HR 1.15; 95% CI 1.05-1.27).
In the patients with active bleeding, however, using filters was tied to a lower risk of mortality at both 30 days (HR 0.68; 95% CI 0.52-0.88) and 90 days (HR 0.73; 95% CI 0.59-0.90).
Filter use did not influence risk of subsequent PE in the first year in any subgroup, but was associated with a greater risk of subsequent DVT in patients without contraindications to anticoagulation (HR 1.53; 95% CI 1.34-1.74) and those with active bleeding (HR 2.35; 95% CI 1.56-3.52).
The reduction in short-term mortality seen in patients with active bleeding who received vena cava filters was unexpected, according to the authors, who note that prior studies showing a potential survival benefit did not account for immortal time bias.
It is possible, they say, that the finding is due to treatment bias stemming from the preferential use of filters in patients who are less sick. They discard that idea, though, because both severity of illness and risk of mortality were higher in patients who received filters in all three subgroups. “Moreover, in the models for death, we adjusted for important risk factors for death, including age, race, insurance status, and the number of comorbidities, in addition to the risk of mortality,” they write.
They also acknowledge the apparent contradiction between the observed reduction in mortality and the lack of an association with recurrent PE.
“If a [vena cava filter] provides a physical barrier that prevents or retards the migration of embolus into the lungs, then [it] should theoretically prevent both early deaths and symptomatic PEs, at least until the time of [filter] retrieval,” they say.
The discrepancy in the current study could be explained by multiple factors, including the fact that anticoagulation therapy could have been restarted in the active bleeding group after the risk of bleeding was considered low enough and the possibility that “a modest proportion” of the filters could have been removed within a few weeks, they say. Also, they add, filter use “might have improved survival not by ‘catching’ large clots that embolized toward the lung, but by breaking up these embolizing thrombi into smaller pieces that were better tolerated hemodynamically.”
White RH, Brunson A, Romano PS, et al. Outcomes after vena cava filter use in non-cancer patients with acute venous thromboembolism. Circulation. 2016;Epub ahead of print.
- The study was supported by the Hibbard E. Williams Endowment at UC Davis.
- White reports no relevant conflicts of interest.