Temporary Vena Cava Filters in Oncology Patients

Background

While inferior vena cava (IVC) filter placements continue to exponentially increase, the long-term complications from these devices are progressively more recognized. Randomized data on the efficacy of filters is sparse and focuses mainly on outcomes following permanent filter placement; however, the majority of filters placed currently are retrievable. Placement and removal of these filters are more expensive than permanent filters and have more long-term complications. In this study, we analyzed the use of retrievable filters in the cancer population, a group at very high risk for incident and recurrent venous thromboembolism (VTE).

Methods

This is a single-institution study. All patients with a history of malignancy or active malignancy that received an interventional radiology (IR) placed temporary IVC filter from 2009 to 2013 were logged into a database. Patients were followed prospectively from time of device placement. Recorded data included demographics, type of malignancy, indication for filter placement, time to filter retrieval, complications of placement/retrieval, rates of VTE recurrence and cause of death (if applicable). Final data analysis (n=179 filter placements) was only performed on patients that had an active malignancy or were receiving adjuvant therapy for a recent active malignancy.

Results

The most common indications cited for filter placement included a contraindication to anticoagulation (69%), surgical prophylaxis (17%) and concern for cardiopulmonary collapse from a pulmonary embolism (PE) (6%). IVC filters were most frequently placed in patients with underlying hematologic malignancies (28%), gastrointestinal malignancies (17%) and gynecologic malignancies (15%). The majority of patients had stage III or IV cancer (61%). Internal medicine providers were most likely to order filter placements (36%) followed by hematologists/oncologists (26%) and gynecologic oncologists (17%). 35% of filters were not placed due to a contraindication to anticoagulation or failure of anticoagulation, and of these filters placed, 20% were not removed. Of the 179 temporary filters placed, 60% remained permanent. The most common reasons stated for failure of filter removal included: progressive disease/clinical deterioration (51%), continued contraindication to anticoagulation (23%) and loss of follow-up (7%). Only 2% of filters were unable to be removed because of mechanical reasons. Of the 81 attempted filter removals, 5 had in-filter thrombus, 4 had surrounding fiber sheaths, 4 had filter tilt, 1 had IVC in-growth, 1 had a procedure related infection and 1 had broken struts. The rate of recurrent VTE in all patients studied was 20% (predominantly deep vein thromboses), with the majority of recurrences occurring in patients that had the filter in place and were not maintained on anticoagulation. By the end of the study, 59% of patients had died, most commonly due to progressive cancer. Median time from filter placement to death was 5.25 months. Additionally, we gathered data on filter costs. Costs were attributable to the device ($1576.00), placement ($10,983.00) and removal ($8,824.00), totaling over $2 million dollars for placement of IVC filters in this cohort.

Conclusions

A significant number of cancer patients who have an IVC filter placed have no contraindication to anticoagulation or evidence of recurrent VTE on anticoagulation. Better prospective data is needed regarding the safety and efficacy of IVC filter placement for prophylactic purposes or in the setting of a large VTE burden as these are commonly cited indications for placement. Additionally, consideration for permanent filter placement should be made in cancer patients as the majority of temporary filters are not removed and may carry higher risks of complications. Notably, our filter removal rate was significantly higher than the retrieval rate at most centers (<20%). IVC filters are commonly placed in patients with advanced malignancy and low expected survival, raising particular questions regarding their role in this patient population. Finally, the cost of filter placement and removal is markedly high, further emphasizing the need for better prospective data to clearly delineate those patients who will derive the most benefit from their use.