SAN DIEGO – Caval angulation and filter tilt complicate removal of inferior vena cava filters, results from a large single-center study demonstrated.
"The use of retrievable IVC filters has increased significantly over the past 10 years," Jennifer Stevens said at the annual meeting of the American Venous Forum. "However, these filters have become permanent in many cases, with retrieval rates ranging anywhere from 15% to 59%. In one multicenter study, the retrieval rate was only 21%."
In an effort to examine patient characteristics and anatomic predictors associated with nonretrieval of IVC filters, Ms. Stevens, a fourth-year medical student at the University of Pittsburgh, and her associates reviewed all retrievable IVC filters that were placed in 404 patients at the university’s medical center between 2004 and 2009. Caval angulation at the lowest renal vein was categorized as straight, less than 30 degrees, 30-60 degrees, and greater than 60 degrees. Filter tilt was classified as none, less than 45 degrees, and 45-90 degrees. Filter thrombus was classified as less than one-third or greater than one-third.
Ms. Stevens reported that the majority of filters (65%) were placed in trauma patients, with prophylactic indications for insertion (57%). More than one-third of patients (41%) were between the ages of 26 and 50 years, and 61% were male.
Of the 404 filters placed, 166 (41%) were not retrieved. "These patients were more likely to be middle-aged males with a therapeutic indication for insertion," she said.
The most common reasons for nonretrieval were physician oversight (38%) and patient noncompliance (20%). Other reasons included significant thrombus that precluded removal (5%), a medical decision to leave the filter permanently placed (11%), and mechanical inability to remove the filter (12%).
The majority of the 24 cases of attempted but failed retrieval were in female trauma patients aged 51-75 years with a history of prior deep vein thrombosis requiring prophylactic filter insertion. "All of these retrievals had more than one separate attempt made before retrieval was aborted," Ms. Stevens said.
Dwell time was higher in patients who had failed retrieval (mean of 67 days vs. 30 days in patients whose filters were successfully retrieved), and 29% were still on warfarin. "Of note, 71% of these patients were categorized with a less than 30-degree angulation at the level of the renal vein," she said.
Ms. Stevens also presented findings from a subset of 39 cases of difficult filter retrievals that required adjunctive maneuvers, most commonly an extra snare (55%). More than half of these cases (62%) involved the Tulip filter.
Of all filters placed, 82 (20%) had apposition of the hook into the caval wall. Of those filters, 9 (11%) were difficult to retrieve.
Predictors of failed or difficult retrieval included any filter tilt or angulation at the renal veins (odds ratio 3.99) and longer dwell time (OR 1.02). Caval penetration by a filter strut was not a significant predictor of failed or difficult retrieval.
"From an anatomic standpoint, caval angulation and filter tilt complicate IVC filter retrieval," Ms. Stevens concluded. "Consideration should be given to deploying in a straight segment of the IVC even if not flush with the renal veins in order to optimize retrieval. Dwell time adversely affects retrieval success, and the overall retrieval rate continues to be moderate, suggesting physician prompts and patient follow-up reminders as possible future targets for improvement."
The study’s principal investigator was Dr. Rabih A. Chaer, a vascular surgeon at the university.
Ms. Stevens and Dr. Chaer had no relevant financial disclosures.