BACKGROUND: Retrievable vena cava filters (RFs) offer the appeal of short-term prophylaxis for trauma patients temporarily at risk for pulmonary embolism (PE) without the long-term risks of permanent vena cava filters (PFs). However, the evidence that RFs and PFs reduce the risks of PE and death in trauma patients is not conclusive. RFs were introduced at our trauma center in August 2002. The purpose of this study was to evaluate the effects of RFs on our strategy to prevent PE in trauma patients. METHODS: We reviewed our trauma registry to compare rates of filter placement, filter-related complications (FRCs), and PE before (Group I: January 2000 to August 15, 2002) and after (Group II: August 16, 2002 through December 2004) RF introduction. Indication for filter placement, filter retrieval, FRCs, and incidence of PE were compared. RESULTS: There were 5,042 patients in Group I and 5,038 patients in Group II. There was a threefold increase in filter placement in Group II compared with Group I (55 [1.1%] versus 161 [3.2%]; p < 0.001). There were no significant differences between the rates of PE (0.2% versus 0.2%, p = 0.636) or major FRCs (1.8% versus 2.5%, p = 0.777). Major FRCs included two filter infections with sepsis, one vena cava thrombotic occlusion, one filter lodged in the jugular vein during retrieval, and one PE after filter placement. RF removal was attempted in 43 (27%) patients and successful in 33 (21%). CONCLUSION: The advent of RFs was associated with a threefold increase in vena cava filter placement in our trauma center. Major FRCs were encountered and a very low incidence of PE was not altered by their use. Successful removal could be verified in only 21% of RFs. The results of this study lead us to question the rationale for a more liberal use of vena cava filters in trauma patients.