Removal of retrievable inferior vena cava filters before discharge: Is it associated with increased incidence of pulmonary embolism?

BACKGROUND: Severely injured trauma patients are at high risk of developing deep venous thrombosis and pulmonary emboli (PE), and may have contraindications to prophylactic or therapeutic anticoagulation. Retrievable inferior vena cava filters (rIVCFs) are used to act as a mechanical obstruction to prevent PE in high risk populations and those with deep venous thrombosis who cannot be anticoagulated. The removal rate of rIVCFs is variable in trauma centers, including our previous published rate of 50% to 89%/year. Indwelling filters carry a risk of significant morbidity and the success of retrieval decreases as the dwell time increases. We hypothesized that once patients could receive appropriate prophylactic or therapeutic anticoagulation, rIVCF could be removed before hospital discharge without impact on occurrence or recurrence of PE. METHODS: All trauma patients with rIVCF placed and removed between January 2006 and August 2018 were reviewed. We collected data from record review from admission to 6 months postfilter removal, including demographics, filter indication, filter type, dwell time, placement and removal complications, antithrombosis medications, location of venous thromboembolism, complications, and discharge disposition. Exposure of interest was timing of filter removal: before (BEF) or after hospital discharge (AFT). The outcome of interest was whether the patient had a documented PE within 6 months of filter removal. RESULTS: A total of 281 rIVCFs were placed, 218 were eligible for removal, 72.4% (158/218) were retrieved with 63% (100/158) removed before discharge. Mean filter duration was 26 days and 103 days for the before and after groups, respectively. No differences (p > 0.05) were noted in the distribution of demographic and clinical factors except for filter indication (venous thromboembolism indication, 95% in AFT vs. 74% in BEF, p = 0.0043). Postremoval PE rates were 0% BEF and 1% AFT (Fisher's exact test, p = 1.000). CONCLUSION: Our results suggest that removal of rIVCFs before discharge once patients are appropriately anticoagulated is a safe strategy to improve retrieval rates. LEVEL OF EVIDENCE: Therapeutic, level V.

Time to stroke: A Western Trauma Association multicenter study of blunt cerebrovascular injuries.

BACKGROUND: Screening for blunt cerebrovascular injuries (BCVIs) in asymptomatic high-risk patients has become routine. To date, the length of this asymptomatic period has not been defined. Determining the time to stroke could impact therapy including earlier initiation of antithrombotics in multiply injured patients. The purpose of this study was to determine the time to stroke in patients with a BCVI-related stroke. We hypothesized that the majority of patients suffer stroke between 24 hours and 72 hours after injury. METHODS: Patients with a BCVI-related stroke from January 2007 to January 2017 from 37 trauma centers were reviewed. RESULTS: During the 10-year study, 492 patients had a BCVI-related stroke; the majority were men (61%), with a median age of 39 years and ISS of 29. Stroke was present at admission in 182 patients (37%) and occurred during an Interventional Radiology procedure in six patients. In the remaining 304 patients, stroke was identified a median of 48 hours after admission: 53 hours in the 144 patients identified by neurologic symptoms and 42 hours in the 160 patients without a neurologic examination and an incidental stroke identified on imaging. Of those patients with neurologic symptoms, 88 (61%) had a stroke within 72 hours, whereas 56 had a stroke after 72 hours; there was a sequential decline in stroke occurrence over the first week. Of the 304 patients who had a stroke after admission, 64 patients (22%) were being treated with antithrombotics when the stroke occurred. CONCLUSIONS: The majority of patients suffer BCVI-related stroke in the first 72 hours after injury. Time to stroke can help inform clinicians about initiation of treatment in the multiply injured patient. LEVEL OF EVIDENCE: Prognostic/Epidemiologic, level III.