De-identified patient information from the State Inpatient Databases (SID) and the States Ambulatory Surgery and Services Databases (SASD) for Florida and California were used to identify all patients who underwent IVC filter placement from 2005 to 2014 and 2005 to 2011, respectively. Hospital practice patterns were assessed as a function of time as well as IVC filter placement and retrieval volume. Temporal trends were evaluated for statistical significance using the Cochran-Armitage test.
A total of 181,260 IVC filters were placed in 178,327 patients over the study period. IVC filter placements peaked in 2010; following the FDA advisory in 2010, however, IVC filter placements monotonically decreased each subsequent year. The proportion of IVCF placement patients with both acute DVT and PE (17.6% vs 11.8%, P < 0.001) at the time of hospitalization increased; likewise, the proportion of IVCF patients with acute DVT or PE with a concomitant acute contraindication to anticoagulation at the time of hospitalization increased as well following 2010 (17.0% vs 11.9%, P < 0.001). From 2005 to 2014, there was a continual increase in both filter retrieval procedures as well overall percentage of filters retrieved. However, estimated retrieval rates remained low, with a retrieval rate of less than 6% in 2014.
Following the FDA warning in 2010, there was a significant decrease in IVC filter placements, with filter placements more frequently performed in patients with poorer health. While retrieval rates increased over time, they remained low.