This article is more than 5 years old. Information may no longer be current.
Vena cava filters after major trauma fail to reduce incidence of pulmonary embolism, death
Click Here to Manage Email Alerts
Placing an inferior vena cava filter in patients within 72 hours of major trauma did not reduce incidence of symptomatic pulmonary embolism or death at 90 days, according to results of a randomized prospective trial published in The New England Journal of Medicine.
“The use of vena cava filters has become widespread in many trauma centers as a primary means to prevent pulmonary embolism in patients who are at high risk for bleeding,” Kwok M. Ho, PhD, intensive care specialist and clinical epidemiologist at Royal Perth Hospital in Australia, and colleagues wrote. “Given the cost and risks associated with a vena cava filter, our data suggest that there is no urgency to insert the filter in patients who can be treated with prophylactic anticoagulation within 7 days after injury.”
Ho and colleagues randomly assigned 240 patients with severe injuries (median injury severity score, 27; range, 0-75) and a contraindication to blood thinners to receive a vena cava filter within 72 hours of injury (n = 122; median age, 35 years; range, 25-56; 77% men) or no filter (n = 118; median age, 41 years; range, 29-58; 76.3% men).
Among the 122 patients in the filter cohort, 89% had the filter installed within 24 hours of enrollment.
A composite of symptomatic PE or death from any cause at 90 days served as the study’s primary endpoint. Symptomatic PE between days 8 and 90 in a subgroup of patients (n = 80) who survived at least 7 days without prophylactic anticoagulation served as a secondary endpoint.
Results — published in NEJM to coincide with the International Society on Thrombosis and Haemostasis 2019 Congress in Melbourne — showed no significant difference in incidence of PE or death among patients who received a vena cava filter compared with those who did not (13.9% vs. 14.4%; HR = 0.99; 95% CI, 0.51-1.94).
However, among the subgroup of patients who did not receive prophylactic anticoagulation within 7 days of injury, five patients — none of whom received a vena cava filter — developed PE, including one who died (RR for PE = 0; 95% CI, 0-0.55).
An entrapped thrombus was discovered in the filter of six patients who received the device.
Researchers noted that the study was underpowered to identify modestly lower occurrence of PE or death in the filter vs. no-filter group, which served as a limitation. They also cited several caveats to the study’s secondary endpoint, including survivor bias because a patient had to survive 7 days to qualify for the subgroup analysis.
“Previous studies have suggested that vena cava filters are associated with an increased risk for deep-vein thrombosis in the legs,” Ho and colleagues wrote. “This complication was not confirmed in this trial — perhaps because of the use of intermittent pneumatic compression in the legs, the initiation of prophylactic anticoagulation, and the removal of the filter as early as possible.” – by John DeRosier
Disclosures: Ho reports advisory roles with Medtronic and research funding from Raine Medical Research Foundation and Royal Perth Hospital. Please see the study for all other authors’ relevant financial disclosures.
Perspective
Back to TopIdo Weinberg, MD, and Yan Hang Fong, MBBS
Inferior vena cava (IVC) filters were developed with the goal of offering an alternative for patients with venous thromboembolism and contraindications to anticoagulation. There is much controversy around the indications of IVC filters, and various societal guidelines differ in their recommendations.
Multiorgan trauma patients are at risk for PE, including fatal PE. Prophylactic IVC filter use has been suggested to prevent PE in these patients, with the intent of reducing mortality. Although this practice may have theoretical merit, data have been limited and risk for IVC filter-related complications may offset any potential benefit of these devices in the absence of appropriate benefit.
Adequately, researchers concluded that early IVC filter placement did not improve patient outcomes. However, several interesting observations should be noted.
Researchers analyzed a prespecified subgroup of patients who survived 7 days, but who did not receive prophylactic anticoagulation within this period due to contraindications (46 in the filter group and 34 in the control group). Although the study was not powered to look at this subgroup, none of the patients in the IVC filter group had symptomatic PE, whereas five (14.7%) in the control group did. Of these, one was a saddle PE diagnosed on postmortem exam. Thus, although widespread use of prophylactic IVC filters in trauma patients is likely wrong, there may be a very specific subpopulation of patients who, if identified beforehand, might benefit from these devices. Unfortunately, identifying such patients, while avoiding overutilization of IVC filters, may prove difficult in clinical practice.
One clue to the nature of these patients was that in the current study, 69% of them had intracranial hematomas or contusions. As a reminder, there is no consensus recommendation from different international societies on the use of prophylactic anticoagulation among patients with intracranial hematomas, likely due to the limited quality of available data.
Also worth mentioning is that in six patients (5% of those who received an IVC filter), retained clots were identified in the IVC filter. This should not immediately denote “dodged PE,” as some of these may have autolysed and one could make the argument that the IVC filter was the cause of these clots. However, perhaps some of these clots signified utility of the device in otherwise sick patients.
Readers should take note of a few additional limitations of the study. First, as the authors recognize, there was an inherent bias introduced regarding decisions on anticoagulation in the filter group. After a filter is placed, the risk for VTE persists, but the contraindication to anticoagulation may have resolved. Therefore, presence of an IVC filter may lead to a delay in starting appropriate anticoagulation.
Second, most patients in both study groups received prophylactic anticoagulation. Although the authors presented an analysis of outcomes among patients who could not receive anticoagulation, these were a relative minority of the whole study. This limits the power of any conclusions about these patients.
Third, timely retrieval (defined as within 90 days), even in the setting of close, dedicated surveillance, was not achieved in 34% of patients. All filters were removed by day 232, except in one patient who became pregnant. Whether this translates to any additional safety concerns was not addressed but has been shown in prior publications. Nonetheless, it reflects the difficulties in daily clinical practice of complying with prompt removal.
Finally, the data cannot be generalized to patients of other ethnicities, especially Asians, for whom thrombotic risk is lower in general, but risk for intracranial bleeding is known to be higher than for Caucasians.
In summary, and in line with most current guidelines, we believe that IVC filters should probably be avoided in most trauma patients. However, we would stop short of stating that they should never be used in this setting. There is likely a subset of trauma patients who can benefit from selective and appropriate IVC filter implantation and retrieval. Hopefully, future studies will better characterize these patients.
Read more about
Click Here to Manage Email Alerts