Endobronchial forceps-assisted complex retrieval of inferior vena cava filters

Objective: The objective of this study was to assess the safety, feasibility, and effectiveness of endobronchial forceps-assisted retrieval of inferior vena cava (IVC) filters when standard retrieval techniques fail. Methods: An Institutional Review Board-approved single-center retrospective review during a 6-year period identified 60 consecutive patients (23 men and 37 women; mean age, 49.3 years; range, 19-77 years) in whom rigid endobronchial forceps were required for IVC filter retrieval after standard techniques failed. Factors affecting retrieval success, including an embedded or tilted filter, overall dwell time, filter fracture, and caval penetration, were recorded, as were success rates, procedural details, and complications. Results: Mean dwell time between filter placement and removal was 565 days (range, 15-7366 days). Various IVC filters were encountered: 33 Option (Argon Medical, Frisco, Tex), 8 Celect (Cook Medical, Bloomington, Ind), 9 Günther Tulip (Cook Medical), 4 G2 and 3 Eclipse (Bard, Murray Hill, NJ), 2 OptEase (Cordis, Bridgewater, NJ), and 1 Simon Nitinol (Bard). Imaging before retrieval showed grade 1 (3/60), grade 2 (44/60), and grade 3 (13/60) filter interaction with IVC wall. The degree of secondary tilt as assessed on images acquired before attempted retrieval was <10 degrees (n = 22; 36.7%), 10 to 20 degrees (n = 26; 43.3%), and >20 degrees (n = 12; 20%), with an average tilt of 13.8 degrees (range, 0-48.9 degrees). Of the 60 filters, 58 were retrieved successfully (96.7%). Two of these required a second attempt (one because of migration into the right atrium and the other because of extreme discomfort of the patient requiring subsequent general anesthesia). Five of 60 procedures were forceps-assisted loop snare retrievals in which the forceps were used to dissect the embedded hook free from surrounding hyperplastic caval venous endothelium. Mean fluoroscopy retrieval time was 33.2 minutes (range, 10-76.9 minutes). Intraprocedural inconsequential filter fracture was observed in 10 patients. There were four complications: one retroperitoneal hemorrhage, one IVC dissection flap, and two filter fractures with subsequent migration of filter components to the right side of the heart and to the right pulmonary artery. The first two complications required hospital admission and conservative management; in the last two, the fractured and migrated filter limbs were retrieved successfully using a snare device. Conclusions: Rigid endobronchial forceps can be safely and reliably used to remove embedded, fractured, or tilted retrievable IVC filters from patients in whom standard retrieval techniques are unsuccessful. There is a high success rate and minimal complications. We propose that the degree of filter tilt, caval penetration, and filter fracture are predictive of the need for the use of forceps as a first-line retrieval technique.