An Algorithm for Inferior Vena Cava Filter Placement in Severely Injured Patients: A Systematic Review-Based Consensus and Guidelines by SMA-VS.

Abstract

Aims/background: Venous thromboembolism is a major complication of severe injury, and the role of inferior vena cava (IVC) filters in patients suffering from this condition remains controversial. The current clinical practice guideline, developed by the Shanghai Medical Association Vascular Surgery Branch, aims to clarify the role of IVC filters in severely injured patients based on evidence and consensus-derived recommendations.

Methods: We conducted a systematic review and random-effects meta-analysis of randomized controlled trials and observational cohort studies, reported in accordance with PRISMA 2020, followed by a two-round Delphi consensus process involving 21 experts from five specialties. Because the included studies spanned clinically heterogeneous populations and combined prophylactic with therapeutic filter placement, a post hoc sensitivity analysis restricted to trauma and fracture populations was performed. Pooled estimates were derived using the DerSimonian-Laird random-effects estimator in Review Manager (RevMan) 5.4. Certainty of evidence was not formally graded; reported agreement percentages denote the strength of expert consensus rather than Grading of Recommendations Assessment Development and Evaluation (GRADE) or equivalent formal ratings. Because each outcome was informed by only two to five studies, pooled estimates should be interpreted with caution.

Results: Six studies were included. In the principal analysis, which pooled studies of both prophylactic and therapeutic filter placement, IVC filters were associated with a lower incidence of symptomatic pulmonary embolism (five studies; risk ratio [RR] 0.22, 95% confidence interval [CI] 0.08-0.66), an association informed by four observational studies and one randomized trial. Filters were not associated with reduced all-cause mortality (RR 0.88, 95% CI 0.41-1.89) or with a difference in bleeding (RR 1.08, 95% CI 0.91-1.28), and were associated with a higher incidence of deep vein thrombosis (RR 1.38, 95% CI 1.02-1.87), although this finding was not robust in leave-one-out analysis. When the analysis was restricted to trauma and fracture populations, the reduction in symptomatic pulmonary embolism persisted (RR 0.15, 95% CI 0.04-0.59), whereas the association with deep vein thrombosis was no longer evident (RR 1.31, 95% CI 0.68-2.50). Fifteen recommendations were formulated, of which fourteen achieved consensus. These recommendations include: (i) pharmacologic thromboprophylaxis is the mainstay of prevention; (ii) IVC filters are reserved for defined indications; and (iii) availability of institutional retrieval infrastructure is a prerequisite for IVC placement. A clinical decision algorithm was developed to integrate the recommendations.

Conclusion: The certainty of evidence was low for most components, although substantial agreement was achieved among expert clinicians. Pharmacologic prophylaxis should remain the default strategy, with prophylactic IVC filter use restricted to selected indications, dependent on retrieval infrastructure, and limited in duration. The seven-day contraindication threshold should be regarded as a deliberative prompt rather than a decision rule and warrants further investigation. Because the available evidence is drawn largely from fracture rather than polytrauma cohorts, and the pooled estimates combine prophylactic with therapeutic filter placement, these recommendations should be applied to severely injured patients with appropriate caution.

Systematic review registration: PROSPERO (CRD420261334942).