Complications Related to Inferior Vena Cava Filters: A Single-Center Experience

We reviewed our experience with the different types of inferior vena cava (IVC) filters used over 4 years for the incidence of complications and correlated this with the type of filter used. This is a retrospective study involving chart reviews of all the patients who received IVC filters placed between January 2002 and January 2006. Data related to indications for filter insertion and the incidence of early (30 days) and late complications related to the filter insertion were collected. Complications were correlated to the type of filter and the indication for insertion. Statistical analysis was done using Fisher's exact test, and p<0.05 was considered significant. During this period 400 filters were inserted. There were 199 males (49.7%) and 201 females (50.25%). The mean patient age was 61 years (range 17-86). Filters used included TrapEase in 224 (56%), Green field filter in 95 (23.8%), Gunther-Tulip in 42 (10.5%), Bard recovery nitinol (all first-generation) in 34 (8.5%), and Simon Nitinol filter in five (1.2%). The indications for IVC filter insertion included acute venous thromboembolism (VTE) event in 273 patients (68.25%) and pulmonary embolism (PE) prophylaxis in 127 (31.75%) patients. In the group with VTE, 59 (21.6%) had contraindication for anticoagulation and 34 (12.5%) had hypercoagulable/malignant conditions. In the 127 patients who received the filter for PE prophylaxis in the absence of VTE, 107 (84.3%) had fractures, 43 (33.9%) had head injury, 32 (25.2%) had multiple trauma, and 15 (11.8%) had paralysis. Sixteen (12.6%) of the prophylaxis patients had IVC filter insertion prior to an elective surgical procedure. Complications in the form of hematoma at the site of filter insertion occurred in four (1%) patients, ipsilateral limb deep vein thrombosis in 15 (3.8%) patients, migration/tilt of filter in six (1.5%) patients, PE in six (1.5%) patients, and IVC thrombosis in 19 (4.75%) patients. Migration/tilt was higher in Bard filters compared to other filters, individually (p<0.004) and as a group (11.8% vs. 0.55%, p<0.0005). All other complication had a comparable incidence in all filters. However, in the group of patients (n=34) who had hypercoagulable/malignant conditions, the incidence of IVC thrombosis was higher with TrapEase filters compared to all other filters as a group (25% vs. 0%, p<0.05). In conclusion, IVC filters are frequently used for prophylaxis in the absence of VTE conditions. Complications are relatively low. All types of filters used in this study had comparable complications with the exception of the Bard filter, which had a higher incidence of tilt, and the TrapEase filter, which had a higher incidence of IVC thrombosis, in patients with hypercoagulable/malignant conditions.