Endovascular embolization for the treatment of epistaxis: Systematic review and meta-analysis

Abstract

Background: Endovascular embolization (EE) is a treatment option for epistaxis refractory to first-line interventions. Data regarding embolization is limited to small case series and a meta-analysis has not been performed.

Methods: PubMed, Scopus, and EMBASE were used to identify studies that reported outcomes for at least 10 patients undergoing EE for epistaxis. Outcomes included procedural success, rebleeding, and complications. Pooled rates for each outcome were obtained with random effects models.

Results: A total of 44 studies comprising 1664 patients met the inclusion criteria. The mean age ranged from 28.1 to 67 years and there were 28.4% females. The pooled procedural success rate was 87% (95% CI 83.9-89.6, I² = 53%). Age (OR 0.95, 95% CI 0.91-1) and hereditary hemorrhagic telangiectasia ([HHT], OR 0.97, 95% CI 0.96-0.99) were associated with decreased odds of success. The pooled rebleeding rate was 16.4% (95% CI 13.6-19.6, I² = 48%), and HHT was associated with greater odds of rebleeding (OR 1.02, 95% CI 1-1.03). The pooled overall complication rate was 14.4% (95% CI 9.8-20.6, I² = 85.8%). The pooled rates of stroke and vision loss were 2.1% (95% CI 1.5-3.1, I² = 1.5%) and 1.8% (95% CI 1.2-2.6, I² = 0%), respectively.

Conclusion: EE for epistaxis has a high rate of procedural success. Interventionalists should be aware of the risk for rebleeding, especially among patients with HHT.

Keywords: Epistaxis; endovascular procedures; therapeutic embolization.

Figure 1.

PRISMA flow diagram demonstrating results of the literature review.

Figure 2.

Forest plot for procedural success showing results from individual studies and the pooled rate calculated from the random effects model.

Figure 3.

Forest plot for rebleeding rate showing results from individual studies and the pooled rate calculated from the random effects model.

Figure 4.

Results from the meta-regression demonstrating the effect of hereditary hemorrhagic telangiectasia (HHT) on procedural success and rebleeding.

Figure 5.

Forest plot for overall complication rate showing results from individual studies and the pooled rate calculated from the random effects model.

Figure 6.

Funnel plots demonstrating publication bias for procedural success, rebleeding, and complication rates. The funnel plots on the right include imputed missing studies obtained with the trim-and-fill method.

Figure 7.

Forest plot comparing the procedural success of endovascular embolization to surgical ligation. (EE + : Number of patients who underwent successful endovascular embolization, EE-: Unsuccessful endovascular embolization, S + : Successful surgery, S-: Unsuccessful surgery).