Abstract No. 222 Twenty Years of Pulmonary Arteriovenous Malformation Embolization: A Meta-Analysis

To determine safety, technical success, and effectiveness of transarterial embolization (TAE) for treatment of pulmonary arteriovenous malformation (PAVM) through meta-analysis of published literature in the last twenty-years. PubMed, Embase, and Cochrane Library databases were queried from June 2003 to June 2023. Studies including >5 patients with PAVM treated with TAE were included. Baseline characteristics, technical success, treatment success/failure, and adverse events were retrieved. Meta-analysis was conducted accounting for overlapping sample sizes. Subgroup analysis was performed based embolization agent and technique. A random-effect model was adopted to achieve a conservative estimate in light of heterogeneity. This meta-analysis included 62 studies. Among 27 studies including 1444 patients that evaluated technical success, the pooled rate was 98.7% (95%CI: 98.0-99.4%). Treatment failure was reported by 35 studies including 2782 patients, yielding a pooled rate of 12.1% (95%CI: 9.1-15.2%). Twelve studies compared outcomes of TAE performed with coils and vascular plugs, Treatment failure occurred in 67/415 PAVMS treated with vascular plugs with or without coils, fewer than 203/661 that were treated with coils-only (odds ratio [OR]: 2.11, 95%CI: 1.14-3.90, p=0.017). However, the advantage of vascular-plug-only over coils-only did not reach statistical significance (169/454 vs 46/170; OR: 1.47, 95%CI: 0.70-3.11, p=0.310). Four studies compared outcomes between feeding artery embolization (FAE) and venous sac embolization (VSE), noting that treatment failure according to size reduction criteria was higher in the FAE group compared to the VSE group (44/181 vs 6/158, OR: 5.34, 95%CI: 2.32-12.27, p< 0.001). The pooled severe adverse event ratio was 0.9% (95%CI: 0.4-1.5%) among 34 studies and 1117 patients. The rate of post-embolization pleuritic chest pain was 16.6% (95%CI: 2.0-21.2%), only one of which required readmission while the remaining were managed conservatively. Post-embolization catastrophic events (brain infection and stroke) were evaluated by 14 studies including 629 patients, reporting a pooled rate of 0.8% (95%CI: 0.0-11.7%). Available literature on PAVM embolization is heterogeneous in PAVM baseline characteristics, definition of treatment success, embolization material/technique, and follow-up criteria. However, TAE has favorable technical and treatment success rates with a low risk for complications. The use of vascular plug was associated with higher treatment success.