Introduction Dural arteriovenous fistulas (dAVF) are relatively rare arteriovenous shunting lesions without intervening nidus.^{1 2} Risk stratification and associated natural history are characterized by the location and nature of venous drainage.^{3 4} Symptoms are related to the location of the recipient vein, such in patients with pulsatile tinnitus with sigmoid sinus drainage, or attributable to venous hypertension secondary to cortical venous drainage. Management options include transarterial or transvenous endovascular routes, surgery, radiosurgery, or observation. Follow-up angiography after treatment is typically performed one year following treatment to ensure complete embolization,⁵ but may be obtained as early as 3 to 6 months in some patients. In an era of advanced imaging techniques such as magnetic resonance angiography and arterial spin labeling, the feasibility of cross-sectional imaging for surveillance of treated dAVFs is investigated in this study.

Methods Patients who underwent endovascular or surgical treatment for a dAVF were extracted from a single institution neurointerventional database. Only patients who received a MRI/MRA with or without ASL between treatment and follow-up catheter angiogram in the past 6 years were included. Demographics and presenting symptoms were recorded. Two fellowship trained neuroradiologists, one also a fellowship trained neurointerventionalist, blinded to patient history independently reviewed the MRI/MRA studies. The readers determined presence of a recurrent or residual dAVF, and subsequently stratified the dAVF into low, medium, or high risk categories based on the Borden classification system. These results were then compared to the angiographic results as interpreted by the operating neurointerventionalists using Pearson correlation coefficients.

Results 13 MRIs obtained following treatment of a dAVF between 2017 and 2023 were evaluated in 8 patients of which 4 were female with mean age 57.8 (range 5-85). The dAVF location was intracranial in 7 patients and cervical in 1 patient. Follow-up with catheter angiography was obtained an average of 73 days following MRI/MRA (range 7-219 days). The inter-reader correlation coefficient was 0.64 (p=0.53). The correlation coefficients between the MRI interpretation and catheter angiogram results were 0.77 (p=0.46) and 0.65 (p=0.53) for each reader.

Conclusion MRI/MRA evidence of residual or recurrent dAVF is not reliably found between two neuroradiologists or as compared to angiographic results. Catheter angiogram remains the standard for follow-up of treated dAVF.

References

1. Newton TH, Cronqvist S. Involvement of dural arteries in intracranial arteriovenous malformations. Radiology. 1969;93(5):1071-1078.

2. Alexander MD, Darflinger R, Cooke DL, Halbach VV. Cerebral arteriovenous fistulae. Handb Clin Neurol. 2021;176:179-198.

3. Chen CJ, Buell TJ, Ding D, et al. Intervention for unruptured high-grade intracranial dural arteriovenous fistulas: a multicenter study. J Neurosurg. 2022;136(4):962-970.

4. Chen CJ, Buell TJ, Ding D, et al. Observation versus intervention for low-grade intracranial dural arteriovenous fistulas. Neurosurgery. 2021;88(6):1111-1120.

5. Abecassis IJ, Meyer RM, Levitt MR, et al. Recurrence after cure in cranial dural arteriovenous fistulas: a collaborative effort by the Consortium for Dural Arteriovenous Fistula Outcomes Research (CONDOR). J Neurosurg. 2022;136(4):981-989.

Disclosures A. Baker: 5; C; Founder, SmartCath Inc.. M. Alexander: None. A. Ye: None. K. Narsinh: None. A. Abla: None. M. Amans: None. D. Cooke: 5; C; Founder, Smartcath Inc. S. Hetts: None.