Indications, Complications and Management of IVC Filters in 952 Patients At An Academic Hospital with a Level 1 Trauma Center

Inferior vena cava (IVC) filters are frequently placed for the treatment and prevention of venous thromboembolism (VTE). Despite limited data to support their safety and efficacy, filter use has increased over the last three decades at our institution, as it has in hospitals throughout the United States. Boston Medical Center (BMC), a 500-bed academic institution with more than 30,000 admissions per year, is the largest and busiest Level 1 Trauma Center in New England. This study investigated the indications for IVC filter placement at BMC and examined outcomes including filter complications, management of anticoagulation, and follow up.

A search was performed after obtaining IRB approval using Current Procedural Technology (CPT) codes to determine the number of patients at BMC that had IVC filters placed from August 2003 to February 2011. IVC filters were placed in 978 patients during the study period. Of these, 26 patients were excluded due to incomplete medical records. The remaining 952 charts were reviewed to determine patient demographics, indication for IVC filter placement, complications, post-discharge follow-up, filter retrieval, and use of anti-coagulation.

Of the 952 filters for which data were available, permanent filters were placed in 273 patients (29%), with the majority of these placed prior to 2006. Retrievable filters were placed in 679 patients (71%). Seven different types of IVC filters were placed by physicians from multiple departments, in decreasing order of frequency: interventional radiology, trauma surgery, vascular surgery, cardiology, and cardiothoracic surgery. Filters were successfully placed in the IVC in all patients except one, in whom a portion of the filter entered the renal vein and could not be extracted. The majority of filters were placed due to perceived immediate contraindication to anticoagulation (recent trauma, bleeding, or surgery). Five hundred and four patients (53%) had VTE at the time of filter placement with almost half of these, 237 patients (25% of all patients), received therapeutic anti-coagulation prior to discharge. A median of three days (range 0–32) elapsed between the date of trauma and filter insertion, with 174 (37%) being inserted five or more days after trauma. Only 31 patients (3%) sustained a VTE while on therapeutic anticoagulation. There was no protocol in place for routine imaging after filter placement or retrieval. Of the patients referred for subsequent imaging (approximately half) due to symptoms of VTE or other indications, 73 had VTE that developed after filter placement. Twenty-six of these events occurred during the index hospitalization. The majority of patients had follow-up at BMC within one month, although 207 patients had no documented follow-up at our institution. An attempt was made to remove 71 (10.5%) of the retrievable filters placed. Retrieval was successful in 58 of the 71 patients. Retrieval attempts failed in thirteen patients (18%) for the following reasons: filter embedded in the IVC, a clot in the filter, or a filter protruding through the blood vessel. The median retrieval time was 122 days (range 2 to 1931 days). The patient with retrieval at 1931 days had a fractured IVC filter removed with one prong extracted from a pulmonary artery.

These data report on the largest single institution review of IVC filter placement and follow-up. Nine hundred and fifty-two filters were placed during the study period and 679 were filters designed for retrieval, yet only 58 retrievable filters (8.5%) were successfully removed. Unsuccessful retrieval occurred in 18% of attempts, making this a common and under-recognized problem. A significant proportion of filters placed for trauma were inserted after the period of highest bleeding risk had subsided, when anticoagulation may have been more appropriate. While many of these filters were placed under the generally accepted guideline of an existing contraindication to anticoagulation, 237 patients were discharged on therapeutic doses of anti-coagulation. Data are critically needed regarding indications for filter placement, risks and benefits of retrieving filters, and short and long-term complications of leaving filters in place.

Sloan:Acetylon Pharmaceuticals: Consultancy; Millenium: Consultancy.