ABSTRACT
OBJECTIVE: The use of inferior vena cava (IVC) filters is controversial. However, the procedure is widely performed for secondary prophylaxis in patients with severe pulmonary embolism (PE), including those treated by a PE response team (PERT). In this study, we analyzed patient factors associated with the clinical decision to place an IVC filter in PERT patients. METHODS: Data were collected on all Massachusetts General Hospital patients who had a PERT activation from October 1, 2012, to January 29, 2019. Data describing demographics, medical history, PE characteristics and treatment were collected at the time of PERT activation and prospectively for one year after PERT activation. Univariate and multivariable regression analyses were performed to determine factors associated with IVC filter placement. RESULTS: We identified 834 patients, of whom 91 (10.9%) had an IVC filter placed in the first 7 days after PERT activation. The majority of patients receiving an IVC filter were male (55/91 [60.4%]; P =.096) with a mean age of 65 ± 15.0 years. Patients who received an IVC filter were less likely to have had a PERT referral from the Emergency Department (ED) (41/544 [7.5%]; P < .001) and more likely to have been referred from the intensive care unit (24/107 [22.43%]; P ≤ .001) compared with a floor referral. Patients who presented with syncope (15/86 [17.4%]; P = .040), a history of recent trauma (12/41 [29.3%]; P < .001), intracranial hemorrhage (11/39 [28.2%]; P = .002), a recent surgery or invasive procedure (30/188 [16.0%]; P = .012), a recent surgery (29/160 [18.1%]; P = .001) and a recent hospitalization (38/250 [15.2%]; P = .009) were more likely to have an IVC filter placed. Patients receiving an IVC filter were also more likely to have evidence of right heart dysfunction on a computed tomography pulmonary angiogram (61/359 [17.0%]; P < .001) and an echocardiogram (26/144 [18.1%]; P = .003). Compared with patients without an IVC filter, the 30-day venous thromboembolism recurrence rate was higher (4.7% vs 11.0%) in patients with IVC filters (10/45 [22.2%]; P = .023). CONCLUSIONS: Factors associated with venous thromboembolism severity (eg, PERT referral from intensive care unit and right ventricular dysfunction) and an increased bleeding risk (eg, recent surgery or trauma) were associated with IVC filter placement among PERT patients.