Sep 062019

Inferior vena cava filters placed prophylactically in patients hospitalized for trauma prevented symptomatic pulmonary embolism in those patients with persistent contraindications to anticoagulation, in a significant randomized trial. However, prophylactic IVC filter placement for all post-trauma patients did not improve outcomes generally.

Trauma teams have always faced a difficult dilemma in the prevention and treatment of deep venous thrombosis (DVT) and pulmonary embolism (PE) in treating hospitalized patients. Almost one in five patients who suffer major trauma experience proximal DVT, and one in ten develop pulmonary embolism. PEs after major trauma are usually not fatal, but still account for about 1 in 8 deaths after successful trauma care. DVTs in trauma patients often develop in the first few days after presentation, and delays in prevention increase the risk of morbidity and mortality.

Because anticoagulation is often considered risky in patients after trauma, inferior vena cava filter insertion is commonly performed as a preventive measure. However, the efficacy of this practice is unknown, and major professional societies’ guidelines disagree in their recommendations. Although currently available inferior cava filters can be removed, they often are not, leading to complications in a fraction of patients. Some have argued the devices are overused. A randomized trial published in the New England Journal of Medicine tested whether the routine practice of IVC filter insertion after trauma reduces the rate of PEs or death.

Authors randomized 240 adults (median age 39) with a contraindication to anticoagulation after major trauma (median injury severity score 27) to receive either a prophylactic removable IVC filter (usually within 24 hours), or not. More than half of patients had head trauma; almost 40% in each group with intracranial pressure monitoring in use. All patients underwent surveillance Doppler ultrasounds of their legs two weeks after enrollment. Prophylactic anticoagulation (e.g., subcutaneous heparin injections) was provided at the discretion of the treating physicians, who were encouraged to do so as soon as feasible. IVC filters were removed after prophylactic anticoagulation was begun, or before 90 days whenever feasible. Patients underwent CT-angiography to diagnose symptomatic pulmonary embolism based on a predetermined protocol.

At 90 days of follow-up, there was no difference between groups in the composite primary endpoint of symptomatic pulmonary embolism and death (13.9% vs. 14.4%). It's important to note that in this analysis, both arms included some patients who received prophylactic chemical anticoagulation, and some who did not, according to their (unblinded) physicians’ discretion.

Patients receiving IVC filters tended to receive more chemical anticoagulation prophylaxis (non-significant statistically). There was no significant difference in bleeding or rates of transfusion between groups. Patients did not have a higher rate of DVT formation observed after IVC filter placement.

Among the pre-specified subgroup of patients not receiving prophylactic anticoagulation in the first 7 days after injury, IVC filters clearly prevented symptomatic pulmonary embolism (zero vs. 14.7%). However, patients had to survive 7 days to be analyzed in this group, injecting some survivorship bias. Because the trial was not blinded, a patient’s assignment could easily have affected the decision whether or not to provide anticoagulation.

There was only one death from pulmonary embolism in the study, which occurred in a control patient without an IVC filter, who had been receiving prophylactic anticoagulation for several days. Interestingly, authors did not report the absolute incidence of symptomatic pulmonary embolism in either group (just the composite endpoint).

IVC filter removal was complex in about 5% of patients, with clot present on the filter or other issues, and one patient required open vascular surgical removal. Almost one-third of IVC filters were left in place at 90 days, mostly due to ongoing illness or contraindications to anticoagulation.

Should IVC filters be placed routinely in trauma patients with contraindications to anticoagulation? This study's findings may do little to sway physicians either way from their current practice. Filters were rarely harmful, and did prevent pulmonary embolism in patients with ongoing contraindications to anticoagulation. The trial was only powered to detect a relatively large 8.5% difference in the primary endpoint; smaller-scale benefits from IVC filters might be present.

On the other hand, most patients without head trauma were able to safely receive anticoagulation prophylaxis within 7 days; in these patients, IVC filters brought no clear benefit, and authors suggested IVC filters should be safely withheld or delayed.

There are likely significant differences between the pathophysiology of DVT and PE after trauma compared with DVT in patients hospitalized with medical problems. Thus, these findings should not be extrapolated to other patient populations. That said, medical ICU teams frequently provide care after fall-related head trauma in elderly patients, whose increased risk for DVT and PE should be recognized.

IVC filters appear to have limited benefits in preventing PE in medical patients, and should not be considered an adequate replacement for anticoagulation. To prevent complications, IVC filters should be removed as soon as feasible in all patients. In the U.S., post-discharge care coordination to arrange removal after IVC filter placement has traditionally been inadequate, but there are some signs that medicolegal pressure may be inducing proceduralists to improve their removal rates.

Source: NEJM

Get our weekly email update, and explore our library of practice updates and review articles.

PulmCCM is an independent publication not affiliated with or endorsed by any organization, society or journal referenced on the website. (Terms of Use | Privacy Policy)


Prophylactic IVC filters prevent PE in high-risk trauma patients, but were often unnecessary