The primary goal of the treatment of cerebral arteriovenous malformations (AVMs) is angiographic occlusion to eliminate future hemorrhage risk. Although multimodal treatment is increasingly used for AVMs, periprocedural hemorrhage after transarterial embolization is a potential endovascular complication that is only partially understood and merits quantification.
Searching the period between 1990 and 2019, the authors of this meta-analysis queried the PubMed and Embase databases for studies reporting periprocedural hemorrhage (within 30 days) after liquid embolization (using cyanoacrylate or ethylene vinyl alcohol copolymer) of AVMs. Random effects meta-analysis was used to evaluate the pooled rate of flow-related hemorrhage (those attributed to alterations in AVM dynamics), technical hemorrhage (those related to procedural complications), and total hemorrhage. Meta-regression was used to analyze the study-level predictors of hemorrhage, including patient age, Spetzler-Martin grade, hemorrhagic presentation, embolysate used, intent of treatment (adjuvant vs curative), associated aneurysms, endovascular angiographic obliteration, year of study publication, and years the procedures were performed.
A total of 98 studies with 8009 patients were included in this analysis, and the mean number of embolization sessions per patient was 1.9. The pooled flow-related and total periprocedural hemorrhage rates were 2.0% (95% CI 1.5%–2.4%) and 2.6% (95% CI 2.1%–3.0%) per procedure and 3.4% (95% CI 2.6%–4.2%) and 4.8% (95% CI 4.0%–5.6%) per patient, respectively. The mortality and morbidity rates associated with hemorrhage were 14.6% and 45.1%, respectively. Subgroup analyses revealed a pooled total hemorrhage rate per procedure of 1.8% (95% CI 1.0%–2.5%) for adjuvant (surgery or radiosurgery) and 4.6% (95% CI 2.8%–6.4%) for curative intent. The treatment of aneurysms (p = 0.04) and larger patient populations (p < 0.001) were significant predictors of a lower hemorrhage rate, whereas curative intent (p = 0.04), angiographic obliteration achieved endovascularly (p = 0.003), and a greater number of embolization sessions (p = 0.03) were significant predictors of a higher hemorrhage rate. There were no significant differences in periprocedural hemorrhage rates according to the years evaluated or the embolysate utilized.
In this study-level meta-analysis, periprocedural hemorrhage was seen after 2.6% of transarterial embolization procedures for cerebral AVMs. The adjuvant use of endovascular embolization, including in the treatment of associated aneurysms and in the presurgical or preradiosurgical setting, was a study-level predictor of significantly lower hemorrhage rates, whereas more aggressive embolization involving curative intent and endovascular angiographic obliteration was a predictor of a significantly higher total hemorrhage rate.