Closed-loop protocol for inferior vena cava filter management significantly improves retrieval: a competing-risk analysis of a 10-year cohort

Résumé fourni par la source

Inferior vena cava (IVC) filters are indicated only in acute venous thromboembolism (VTE) with an absolute contraindication to anticoagulation and should be retrieved once anticoagulation can be resumed. However, real-world retrieval rates remain low and highly variable. The objective is to assess whether implementing an institutional protocol improved appropriateness of IVC-filter placement and retrieval rates at 90 and 180 days, and to evaluate safety outcomes. Retrospective pre-post single-center study including adults who received an IVC filter between January 2015 and July 2025. Co-primary outcomes were appropriateness at placement (acute VTE ≤ 30 days plus absolute contraindication to anticoagulation) and retrieval, evaluated at 90 and 180 days using Fine-Gray competing-risk models. Among 234 patients (pre n = 182; post n = 52), cancer prevalence was similar (40.1% vs 38.5%). Appropriateness increased from 74.7% to 88.5% (risk ratio 1.18, 95% CI 1.04-1.35; p = 0.036). Retrieval improved at day 90 (41.2% → 65.4%, p = 0.020) and at day 180 (41.8% → 73.1%, p < 0.001). In Fine-Gray models, the post-protocol period was associated with a higher cumulative incidence of retrieval (adjusted sub-HR 1.68 [1.16-2.43] at 90 days; 1.92 [1.39-2.66] at 180 days), consistent with cause-specific Cox estimates. Procedure-related complications did not increase. Although overall mortality was unchanged, retrieval was independently associated with lower 180 days (adjusted HR 0.20 [0.08-0.48]). Implementing a closed-loop, guideline-aligned protocol improved appropriateness and substantially increased IVC-filter retrieval without compromising safety. Older age and active cancer remained barriers to retrieval, highlighting priority groups for targeted follow-up strategies.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.