PURPOSE: To investigate the long-term outcomes of using vena cava filters to prevent symptomatic pulmonary embolism (PE) in major trauma patients who have contraindications to prophylactic anticoagulation. METHODS: This was an a priori sub-study of a randomized controlled trial (RCT) involving long-term outcome data of 223 patients who were enrolled in Western Australia. State-wide clinical information system, radiology database and death registry were used to assess long-term outcomes, including incidences of venous thromboembolism, venous injury and mortality beyond day-90 follow-up. RESULTS: The median follow-up time of 198 patients (89%) who survived beyond 90 days was 65 months (interquartile range 59-73). Ten patients (5.1%) died after day-90 follow-up; and four patients developed venous thromboembolism, including two with symptomatic PE, all allocated to the control group (0 vs 4%, p = 0.043). Inferior vena cava injuries were not recorded in any patients. The mean total hospitalization cost, including the costs of the filter and its insertion and removal, to prevent one short- or long-term symptomatic PE was A$284,820 (€193,678) when all enrolled patients were considered. The number of patients needed to treat (NNT = 5) and total hospitalization cost to prevent one symptomatic PE (A$1,205 or €820) were, however, substantially lower when the filter was used only for patients who could not be anticoagulated within seven days of injury. CONCLUSION: Long-term complications related to retrievable filters were rare, and the cost of using filters to prevent symptomatic PE was acceptable when restricted to those who could not be anticoagulated within seven days of severe injury.