There are sparse data to support the recommendation for inferior vena cava (IVC) filters in patients with recurrent pulmonary embolism while on anticoagulant therapy.

This was a retrospective cohort study of administrative data from the Premier Healthcare Database, 2009-2014. All-cause mortality according to the use of IVC filters was evaluated in patients who suffered a recurrent pulmonary embolism within 3 months of an index pulmonary embolism. Patients were identified by International Classification of Disease, 9th Clinical Modification codes. A time-dependent analysis controlled for immortal time bias.

An IVC filter was inserted in 603 of 814 (74.1%) of patients hospitalized for recurrent pulmonary embolism within 3 months of an index pulmonary embolism. Mortality with an IVC filter was 18 of 603 (3.0%) vs 83 of 211 (39.3%) (P < .0001) without a filter. Among patients with recurrent pulmonary embolism who were stable and did not receive thrombolytic therapy or undergo pulmonary embolectomy, mortality with an IVC filter was 15 of 572 (2.6%) vs 72 of 169 (42.6%) (P < .0001) without a filter.

In the United States, usual practice was to insert an IVC filter in patients with early recurrent pulmonary embolism. Mortality was lower in those who received an IVC filter. Even stable patients with early recurrent pulmonary embolism showed a decreased mortality with IVC filters, even though in other circumstances, IVC filters do not reduce mortality in stable patients. Additional cohort studies would be useful in the absence of a randomized controlled trial.