Background: Inferior vena cava (IVC) filters are indicated in patients with deep vein thrombosis (DVT) or pulmonary embolism (PE) who either have contraindications to or have failed anticoagulation (AC). Given rising concerns about their safety and efficacy, the FDA has issued a communication urging clinicians to remove filters (optimally, within 90 days post-implantation). According to national data retrieval rates remain low. Our study aimed to assess IVC filter retrieval rates and factors that influence retrieval.
Methods: This is a single center, retrospective cohort study of patients who had IVC filter placement between December 2015 and December 2018. Subjects were identified using procedural codes for IVC filter insertion; data regarding demographics, comorbidities, retrieval, IVC filter-related complications and subsequent thromboembolic events were obtained by direct chart review. Survival analyses and a Cox regression model were performed using JMP statistical software.
Results: Over 3 years, 494 patients with IVC filters were identified; 305 (62%) were retrievable. The average age at placement was 69±16 years; 249 (50%) were men and 332 (67%) were Caucasian. After excluding patients who died or were lost to follow-up within 30 days of placement or were discharged to hospice from the index admission, 258 patients with retrievable filters remained (54 retrieved). Indications for IVC filter placement were PE ± DVT 90 (35%), proximal DVT 159 (62%) and prophylactic 9 (3%). Forty two percent of patients (109) were restarted on AC at discharge, while an additional 18% (total 155) received AC at some point thereafter. The rate of retrieval was 8% at 90 days, 23% at 1 year and 28% at 2 years (Figure A). The proportional hazards model identified resumption of AC at any time (HR 3.11, 95%CI 1.6-6.8, p=0.0006) as the strongest predictor of retrieval; AC at discharge was not predictive. Advanced age at placement (HR 0.97 per unit change, 0.96-0.99, p=0.004) and active malignancy (HR 0.5, 95%CI 0.24-0.98, p=0.04) were associated with a lower likelihood of retrieval. The initial thrombotic event, the reversibility of the contraindication to AC, the placing service, sex, ethnicity and other comorbid conditions did not have an impact on retrieval. Kaplan-Meier analysis revealed that subjects who ever resumed AC had significantly higher rates of retrieval at 90 days (11% vs 3.4%) and at 1 year (33% vs 9.7%, log-rank p=0.0003, Figure B) when compared to those who did not. Only four patients experienced IVC filter-related complications (2 filter thrombosis, 1 IVC penetration, 1 device tilting); all occurred 2 or more years after placement. Recurrent thromboembolic events occurred in 50 patients (5 PE, 48 proximal DVT) with no significant difference in frequency between subjects with retrieved and non-retrieved filters; one PE and one DVT occurred at 1 month and 1 week respectively after retrieval.
Conclusion: Despite efforts to increase awareness of IVC filter-associated complications, the unweighted retrieval rate remained below the nationally reported average of 30%. Persistent risk factors for thrombosis such as active malignancy or increasing age and poor prognosis may play a role in the decision to defer retrieval. In our study, resumption of AC proved a powerful predictor of retrieval, with rates approaching expected values in this population. Active surveillance for resolution of contraindications to AC post-IVC filter placement is crucial in increasing retrieval rates.
Figure
Disclosures
No relevant conflicts of interest to declare.
Practice patterns of retrievable inferior vena cava filters and predictors of filter retrieval in patients with pulmonary embolism