Intracranial Dural Arteriovenous Fistulas With and Without Pial Artery Supply: Analysis of Treatment Outcomes

The prevalence of pial arterial supply (PAS) to intracranial dural arteriovenous fistulas (DAVFs) and its implications for the management of these fistulas have been limited to relatively small cohort studies and remain somewhat controversial. We conducted a retrospective study to characterize PAS in DAVFs and explore its implications for treatment.

Consecutive patients evaluated over a 21-year period were retrospectively reviewed. Angiograms were examined to characterize the angioarchitecture of DAVFs and identify the presence of PAS. PAS was classified into 2 types: dilated preexisting dural branches and pure pial supply. Baseline characteristics, treatment approaches, and treatment and follow-up outcomes were compared between the DAVF cohorts with and without PAS. To minimize patient selection bias, the 2 cohorts were matched in a 1:1 ratio using propensity score matching.

In this cohort, 259 out of 1101 patients (23.5%) exhibited an additional PAS. Multivariate analysis identified 7 independent predictors of PAS: younger age ( P < .001), longer disease duration ( P = .021), multiple DAVFs ( P < .001), tentorial DAVFs ( P < .001), transverse-sigmoid sinus DAVFs ( P < .001), and the presence of venous ectasia ( P = .019) and congestion ( P < .001). Complication rates were higher in the PAS group, particularly for postoperative hemorrhage ( P < .001) and ischemia-related complications ( P < .001), which remained significant even after propensity score matching ( P = .013 and P = .001).

The findings suggest that embolization of PAS before DAVF closure may significantly increase the risk of both intracranial hemorrhagic and ischemic complications. Therefore, routine embolization of PAS before DAVF closure is not supported by these results, particularly given the exceptionally low incidence of presumed hemorrhagic complications arising from an unobliterated "pure" pial supply before DAVF obliteration.