Stereotactic radiosurgery for intracranial dural arteriovenous fistulas: patient outcomes and lessons learned over a 3-decade single-center experience
Pierce A Peters,Ryan M Naylor,Giuseppe Lanzino,Michael J Link,Bruce E Pollock
DOI: https://doi.org/10.3171/2024.6.JNS24547
2024-08-30
Abstract:Objective: The role of stereotactic radiosurgery (SRS) in the management of intracranial dural arteriovenous fistula (dAVF) is unclear given the rarity of this lesion and the variability in treatment paradigms. This study describes a 3-decade experience with the SRS technique and its outcomes for patients with dAVF. Methods: The authors conducted a retrospective analysis of patients with dAVF who had undergone single-fraction SRS in the period from 1990 to 2021. The imaging modality initially used for targeting was angiography alone, then angiography plus MRI, and most recently MRI alone. Results: Two hundred twenty-two patients underwent SRS alone (n = 56, 25%) or SRS plus embolization (n = 166, 75%), depending on the severity of symptoms or the presence of cortical venous drainage (CVD). Most patients were women (64%), and the median patient age was 60 years. Common presenting symptoms were pulsatile bruit (55%), visual change or chemosis (21%), headache (10%), and intracerebral hemorrhage (5%). The most frequent dAVF location was the transverse or sigmoid sinus (44%), followed by the cavernous sinus (24%), jugular bulb (9%), and torcula (5%). CVD was noted in 28% of cases, and venous ectasia in 5%. Borden dAVF types among the patients were I (72%), II (20%), and III (8%). Cognard dAVF types among the patients were I (44%), IIa (27%), IIb (5%), IIa+b (15%), III (4%), and IV (5%). The median SRS treatment volume was 7.6 cm3; the median margin and maximum doses were 18 and 36 Gy, respectively. Follow-up after SRS was available for 209 patients (median follow-up 31 months). Obliteration was noted in 75% of the patients (110/147) with follow-up vascular imaging; the median time to obliteration was 37 months. Multivariate analysis revealed that a cavernous sinus dAVF location was predictive of radiological obliteration (HR 1.86, 95% CI 1.08-3.18, p = 0.024). The absence of CVD was predictive of obliteration in subgroup analysis of non-cavernous sinus dAVF (HR 0.53, 95% CI 0.29-0.98, p = 0.04). Symptoms resolved in 86% of patients (160/185) with clinical follow-up. Twelve patients (5.4%) had complications related to angiography for SRS planning (n = 2, 0.9%), embolization (n = 3, 1.4%), post-SRS hemorrhage (n = 1, 0.5%), delayed sinus thrombosis (n = 1, 0.5%), radiation-induced tumors (n = 2, 0.9%), and chronic encapsulated expanding hematoma (n = 3, 1.4%). Conclusions: SRS alone or in conjunction with embolization provided obliteration and symptom relief for the majority of patients with dAVF, with a low rate of procedure-related morbidity. Patients are at risk for late radiation-related complications, which can require treatment many years after SRS.