Background Spinal arteriovenous fistulas can be treated either by surgery or by endovascular means, using different strategies. The main drawback of embolization is the risk of recurrence. Our objective is to evaluate the angiographic occlusion rate and the predictive factors of angiographic cure of spinal arteriovenous fistulas at 3 months or more after embolization.

Methods This is a retrospective single-center study including 38 consecutive patients with spinal arteriovenous fistulas treated by embolization as first-line treatment. We reviewed clinical and imaging data, complications, and the immediate angiographic occlusion rate of the fistulas, and at 3 months or more after the embolization.

Results A total of 45 embolization procedures were performed: 30 procedures using glue, 15 using Onyx by ‘pressure cooker’ or ‘balloon pressure’ techniques. We observed no statistically significant difference between the two groups concerning the immediate angiographic occlusion rate (87% in both groups; P>0.9), as well as for periprocedural complication rates. The angiographic occlusion rate at 3 months or more was higher in the Onyx ‘combined’ techniques treated group (87% vs 40%, P=0.007). The use of Onyx ‘combined’ techniques was independently associated with angiographic cure at 3 months after embolization (P=0.029). No other factors were identified as predictive of angiographic cure and clinical recovery after embolization procedures, nor were any predictive factors identified for the occurrence of periprocedural complications.

Conclusion Embolization of spinal arteriovenous fistulas with Onyx using ‘combined’ techniques appears to be safe and associated with a higher rate of angiographic occlusion at 3 months than regular embolization with glue.