Back to 2017 Annual Meeting Posters

Background: Venous thromboembolism (VTE) results in significant annual morbidity and mortality. Placement of inferior vena cava (IVC) filters has seen a rise in recent years as reported in recent literature with no demonstrated increase in VTE rates as noted by literature. Placement of IVC filters has become commonplace with laxity in the indication guidelines across multiple specialties. In an effort to reduce egregious filter insertion rates, we implemented a strict indication policy. We reviewed the patterns of IVC filter placement before and after implementation of the policy.
Methods: Within a single, 715 bed tertiary care hospital, all charts of patients receiving an IVC filter during a 6 year period (2010-2016) were reviewed. We compared patient demographics, co-morbidities, length of stay, filter types, interventionalist performing procedure and indication for placement. The patterns of filter placement were reviewed before and after implementation of a strict indication policy implemented at the end of 2012. We compared information over a 3 year period per group. Statistical analysis was performed utilizing chi-square and paired T-testing.
Results: A total 782 filters were placed between 2010-2016. No procedural related complications were noted. There were 383 females and 399 males and mean patient age was 70.9 years. The average length of stay was 16.1 days. Prior to policy implementation the average number of filters placed were 167 per year while the average number placed following the policy was 100 filters per year, a 41% reduction (p=.02). Prior to the policy 55% (277) filters were placed for absolute indications, 30% (150) relative and 10% (50) prophylactic. Following implementation of policy 74% (209) were placed for absolute indication 19% (54) relative and 6% (18) prophylactic. Vascular surgeons placed filters for absolute indication 66% (217) of the time and relative/prophylactic 34% (182) prior to the policy. Post-policy, vascular surgeons placed 84% (182) filters for absolute indications and 16% (38) as relative/prophylactic. Non-vascular surgeons placed filters 39% (61) for absolute indications and 61% (93) for relative/prophylactic prior to the policy. Post-policy, non-vascular surgeons placed 46% (27) for absolute indications and 54% (34). Filters placed by vascular surgeons prior to policy was only 63% of the time, following implementation of policy, vascular surgeons placed over 84% of filters (p=.09). Filter types included 16% non-retrievable and 84% (631) retrievable. Although a strong trend was noted towards removable filter placement only 8% were actually retrieved. No statistical significance was demonstrated in rates of pulmonary embolus in the two group comparison (p=.48).
Conclusions: Implementation of a strict indication policy significantly reduced the number of unnecessary IVC filters. Additionally, vascular surgeons placed significantly more filters for absolute indication compared to non-vascular surgeons. Finally, we saw no statistical or clinically significant difference in pulmonary embolus rates when comparing pre-policy with post-policy patients despite placing fewer filters post-policy. It remains the duty of all physicians to carefully select patients that require an IVC filter; a strict indication policy for IVC filter placement can significantly reduce the number of unnecessary filters.