Does timely removal of retrievable inferior vena cava filters improve the rate of device-related adverse events?

Abstract

Patients who underwent both filter placement and retrieval at our institution were included. Chart review was used to obtain relevant data including indication for filter placement, length of the removal procedure, retrieval success, and the use of advanced retrieval techniques. Results: 196 filter placement / retrieval pairs were identified involving 192 patients (93 men, 99 women, mean age 57 Ϯ 14.6 years). Filter types were Denali (107), Option (43), Gunther Tulip (39), Celect (6), and Greenfield (1). The most common indications for filter placement included pre-operative or post-operative thromboembolic disease (61%), history of hemorrhage on anticoagulation or other risk factors for bleeding (21%), and massive pulmonary embolism (5%). Average length of the retrieval procedure was 34.3 minutes with Option retrievals requiring significantly more time (48.8 min; po0.01) when compared to the rest (Tulip 34.8 min, Denali 28.8 min, and Celect 25.5 min). IVC filter retrieval was unsuccessful in 8 of the 196 cases (4%) including 5 Option, 2 Tulip, and 1 Denali filters. Retrieval failure was significantly higher for the Option filters (11.6%; po0.01). Advanced retrieval techniques (triple snare, snare over loop guidewire, or forceps) were required for 19 of the retrievals (9.6%). The use of advanced techniques was significantly higher with the Option filters (32.6%; po0.01). Complications arose during 4 of the retrieval procedures, including access site hematoma, induced arrhythmia, hemopericardium, and extravasation from the inferior vena cava. Conclusions: IVC filters can be safely removed in the majority of patients particularly when the operators are familiar with advanced retrieval techniques. Of the types included in the study, the Denali and Gunther Tulip filters had comparable retrieval statistics while Option filters were more difficult to remove based on multiple metrics.