Outcome of Endovascular Therapy Aiming for Single-session Obliteration of Intracranial Dural Arteriovenous Fistulas

Abstract

The goal of dural arteriovenous fistula (dAVF) treatment is obliteration of the arteriovenous shunt and/or retrograde leptomeningeal venous drainage (RLVD). Single-session obliteration could improve symptoms early and reduce risk of neurological sequelae. This study investigated the efficacy and adverse events of endovascular therapy (EVT) aiming for single-session obliteration in dAVF treatment. We retrospectively examined post-treatment arteriovenous shunt status, number of treatments per case, treatment-related complications, and long-term outcome in 92 dAVF patients who underwent initial EVT at our institution. Single-session obliteration was intended in all cases, but a second session was performed in cases of partial shunt occlusion or remaining RLVD. Complete occlusion was achieved in 85 cases (92.4%) after the single session; RLVD was obliterated in 66 of the 67 Borden type II and III cases combined (98.5%). A second session was necessary in seven cases (7.6%). Complete shunt obliteration was eventually achieved in all cases. The average number of treatments was 1.08 per case. dAVF-related stroke and mortality did not occur after the treatment. On the other hand, radiation-induced skin erythema and alopecia, although all symptoms were transient, occurred in 26 cases (28.3%). Over an average 60.2-month follow-up period, recurrence was observed in seven cases (7.6%). Single-session obliteration was successful in 92% of cases. Especially, single-session obliteration of RLVD may contribute to early prevent of future stroke events. However, reducing total radiation dose during each session is an issue of further study.

Keywords: dural arteriovenous fistula; embolization; endovascular; radiation; session.

Fig. 1

Pre- and post-treatment (just after single-session) digital subtraction angiography (DSA) showing classification of treatment outcome: class 1, complete occlusion, class 2, slight shunt remnant (arrow), class 3a, remarkable shunt remnant (partial occlusion), and class 3b, retrograde leptomeningeal venous drainage (RLVD) remnant (arrows) on post-treatment DSA.

Fig. 2

A scatter plot showing all patient’s operation time (minute), air kerma (mGy), and presence (black)/absence (gray) of their skin disorder. The operation time and radiation dose showed a positive correlation (Pearson correlation coefficient was 0.875). The patients who suffered skin disorders had significantly longer operation time (p <0.001) and higher air kerma (p <0.001) compared with those without skin disorders.