Long-term Outcomes after Staged and Multimodal Treatment for Diffuse Intracranial Dural Arteriovenous Fistulas Involving Interconnected Dural Sinuses: A Single-center Case Series

Abstract

Diffuse intracranial dural arteriovenous fistulas involving interconnected dural sinuses are uncommon lesions with complex angioarchitecture and an evolving clinical course. However, long-term treatment outcomes and surveillance strategies remain poorly characterized. We retrospectively reviewed 49 consecutive patients with intracranial dural arteriovenous fistulas treated at our institution for 10 years. The study included 9 patients with diffuse intracranial dural arteriovenous fistulas involving 2 or more interconnected dural sinus compartments. Endovascular treatment was used as the first-line therapy for symptomatic or hemodynamically significant lesions, and stereotactic radiosurgery was incorporated for residual low-flow shunts or selected lesions that were not amenable to safe or complete endovascular treatment. Follow-up magnetic resonance imaging and digital subtraction angiography were used to assess shunt occlusion, recurrence, and newly detected shunt lesion in other dural sinus compartments. Nine patients were male. On a sinus-compartment basis, 30 shunt-related sinuses were identified. Shunting most commonly involved the transverse-sigmoid sinus, followed by the torcular confluence and the superior sagittal sinus. Cortical venous reflux was observed in 6 patients. Five patients showed multi-compartment shunt involvement at diagnosis, whereas 4 indicated newly detected shunt involvement during follow-up. Staged multimodal treatment achieved complete occlusion in 18 shunting compartments and partial occlusion in 8 with improved venous drainage; the remaining 4 were managed conservatively. The mean follow-up was 49 months (range, 20-70 months), and patients had favorable outcomes with a modified Rankin Scale score of 0-2. Diffuse intracranial dural arteriovenous fistulas may follow an evolving angiographic course even after treatment, underscoring the importance of long-term follow-up and individualized staged management.

Introduction

Dural arteriovenous fistulas (dAVFs) are abnormal arteriovenous shunts between the dural arteries and dural sinuses and/or cortical veins, accounting for approximately 10%-15% of intracranial vascular malformations.¹⁾ The natural history of dAVFs largely depends on venous drainage patterns. Lesions with cortical venous reflux (CVR) are considered aggressive because of their strong association with intracranial hemorrhage and progressive neurological deficits.²⁾

A subset of intracranial dAVFs exhibits a diffuse shunt distribution across multiple interconnected dural sinus compartments. These lesions are frequently associated with venous hypertension, sinus stenosis or occlusion, and a complex angioarchitecture. Therefore, achieving a definitive cure using a single procedure is often difficult. In addition, angiographic changes observed during follow-up, including recurrent shunting or newly detected shunt involvement in other sinus compartments, may further complicate management.^3-6)

This study aimed to describe the clinical characteristics and treatment outcomes in patients with diffuse intracranial dAVFs involving interconnected dural sinuses treated at our institution and highlight practical considerations for long-term management.

Case Reports

Study design and patients

This single-center retrospective observational study was approved by the institutional review board of our hospital (approval number O-1881). We reviewed 49 consecutive patients with intracranial dAVFs who underwent treatment at our institution from April 2015 to March 2025. Of 49 patients, 9 (18%) with diffuse intracranial dAVFs involving 2 or more interconnected dural sinus compartments were included in this study. The shunt distribution was assessed on a sinusoidal basis for descriptive purposes, including the transverse sigmoid sinus (TSS), torcular confluence, superior sagittal sinus (SSS), and jugular bulb (JB). For clinical purposes, cases were categorized into those showing multi-compartment shunt involvement at the initial diagnosis and those showing newly detected shunt involvement in another dural venous sinus compartment during follow-up. The newly detected shunt was defined as shunt involvement not identifiable on the initial angiographic study but subsequently detected in a different dural sinus compartment on comparative review of serial angiography.

Angiographic examination and definitions

All patients underwent digital subtraction angiography (DSA) for initial diagnostic examination and post-treatment assessment. Angiographic characteristics, including the distribution of shunts across the dural sinus compartments, arterial feeders, and venous drainage patterns, were assessed by a serial comparative review of DSA, with particular attention paid to the presence of CVR and classification according to the Cognard system.²⁾ We also evaluated the presence of venous varices and pseudophlebitic pattern (PPP), defined as engorged and tortuous cortical veins with delayed emptying in the venous phase, typically accompanied by stenosis or occlusion of dural venous sinuses.⁷⁾ Serial angiographic comparisons were also used to assess interval changes in shunt distribution across the dural venous sinus compartments.

Treatment strategy

Treatment strategies were tailored to each individual patient, with priority given to hemodynamic stabilization, improvement of venous drainage, and reduction of CVR rather than immediate angiographic obliteration of all shunts. Therapeutic interventions are generally indicated for symptomatic or hemodynamically significant lesions, including those associated with CVR, intracranial hemorrhage, progressive cognitive decrease, papilledema, venous hypertension, and substantial shunt flow. Endovascular treatment (EVT) is the first-line treatment for lesions considered amenable to endovascular access, particularly for high-flow shunts requiring prompt flow reduction or improvement of venous drainage. Depending on the angioarchitecture of each lesion, patients underwent transarterial embolization (TAE), transvenous embolization (TVE), or a combination of both. Liquid embolic agents such as N-butyl cyanoacrylate (NBCA) or Onyx were used for TAE, whereas detachable coils were used for TVE. Stereotactic radiosurgery (SRS) was incorporated in a staged manner for residual low-flow shunts, lesions not amenable to safe or complete EVT, and selected anatomically complex lesions in which preservation of normal venous drainage was considered important, particularly those centered on the torcular confluence. SRS is also used as a primary treatment option for selected small low-flow lesions. SRS was delivered using a Gamma Knife unit with a marginal dose of 18-20 Gy and a maximum dose of 36-40 Gy per lesion. In contrast, asymptomatic low-grade lesions without CVR or stable venous drainage were managed conservatively with close follow-up imaging.

Follow-up and outcomes

After treatment, patients underwent follow-up magnetic resonance imaging (MRI) 3 months later and every 6 months thereafter. When recurrence or newly detected shunt lesion was suspected during the follow-up, DSA was performed for further evaluation. Follow-up imaging was used to assess shunt status, including complete occlusion, recurrence, and newly detected shunt involvement in another dural sinus compartment. Complete occlusion was defined as complete disappearance of the shunt flow on DSA. Recurrence was defined as the reappearance of shunt flow in a previously treated, angiographically occluded compartment. Newly detected shunt lesion was defined as shunt identified in a dural venous sinus compartment that was undetectable in the initial angiographic study.

Summary of the series

All 9 patients were male, with a mean age of 63.4 years (Table 1). Pulsatile tinnitus was present in all patients at presentation, and some patients had additional symptoms, including cognitive decrease (3 patients), papilledema and/or visual disturbance (2 patients), intracranial hemorrhage (3 patients), and seizures (4 patients); several patients exhibited multiple manifestations. On a sinusoidal basis, 30 shunt-related sinuses were documented in 9 patients. Shunt involvement was observed in the TSS in 12 compartments, at the torcular confluence in 8, at the SSS in 7, and in the JB region in 3. Notably, 8 of the 9 patients had shunt involvement including the torcular confluence. Shunts tended to be widely distributed, and 8 patients had the involvement of 3 or more sinus compartments during follow-up. The mean number of arterial feeders per shunt compartment on initial DSA was 6. CVR was observed in 6 patients, all of whom also showed PPP.

Table 1

Diffuse Intracranial Dural Arteriovenous Fistulas (9 patients, 30 lesions): Baseline and Angiographic Characteristics

Case No.	Age (yrs) /sex	Sites of lesions	Type	Symptoms	ICH	Epilepsy	Cognard type	Feeders	CVR	Deep drainage	Varix	PPP	sinus stenosis/occlusion
APhA: ascending pharyngeal artery; BVR: basal vein of Rosenthal; CVR: cortical venous reflux; ICA: internal carotid artery; ICH: intracerebral hemorrhage; JB: jugular bulb; Lt.: left; M: male; MMA: middle meningeal artery; OA: occipital artery; PAA: posterior auricular artery; PCA: posterior cerebral artery; PMA: posterior meningeal artery; PPP: pseudophlebitic pattern; Rt.: right; SCA: superior cerebellar artery; SPS: superior petrosal sinus; SSS: superior sagittal sinus; STA: superficial temporal artery; StrS: straight sinus; TSS: transverse sigmoid sinus *de novo.
1	56/M	Lt. TSS, Confluence, SSS, Lt. JB	Synchronous	Tinnitus	−	−	II a+b	OA, APhA, PAA, MMA, STA, PMA	+	−	−	+	Lt. JV
2	63/M	bilateral TSS, Confluence	Synchronous	Tinnitus, Papilledema	+	−	II b(Lt.TSS) II a+b	OA, APhA, PAA, MMA, STA	+	StrS	+	+	Lt. TS
3	78/M	Lt. TSS, Confluence, SSS	Synchronous	Tinnitus, Dementia, Papilledema	−	+	II a+b	OA, APhA, PAA, MMA, STA, PMA, ICA	+	−	−	+	−
4	61/M	Lt. TSS, Lt. JB	Synchronous	Tinnitus	−	−	II a	OA, APhA, PAA, MMA, STA	−	SPS	−	−	−
5	73/M	Lt. TSS, Confluence, SSS	Synchronous	Tinnitus	−	−	II a	OA, APhA, PAA, MMA, STA, ICA	−	StrS	−	−	Rt. TS, SSS
6	26/M	SSS, Lt. TSS*, Confluence*	Metachronous	Tinnitus	−	+	IV (SSS) II a+b	OA, APhA, PAA, MMA, STA, PMA	+	BVR	+	+	SSS
7	73/M	Rt. TSS, Confluence, SSS, Lt. TS*	Metachronous	Tinnitus, Dementia	+	+	II a+b II a(Lt. TS*)	OA, APhA, PAA, MMA, STA, Thyrocervical artery, ICA, PCA, SCA	+	StrS	−	+	−
8	70/M	Rt. TSS, Confluence, SSS*	Metachronous	Tinnitus, Dementia	−	−	II a+b	OA, APhA, PAA, MMA, STA, PMA, ICA	−	StrS	−	−	−
9	71/M	Lt. TSS, Confluence, SSS, Rt. TSS*, Lt. JB*	Metachronous	Tinnitus	+	+	II a+b	OA, APhA, PAA, MMA, STA, PMA, ICA	+	StrS	+	+	SSS

Multi-compartment shunt lesions were indicated in 5 at the initial diagnostic examination, whereas in the remaining 4 patients (44%), a newly detected shunt lesion was shown in another dural sinus compartment during follow-up. In 2 of these 4 patients, newly detected shunt lesions appeared after complete occlusion of the initially treated compartments had been achieved. The interval between treatment and the detection of new lesions ranged from 3 to 31 months. In addition, recurrent shunt flow in previously treated and angiographically occluded compartments was observed in 3 patients, including 2 after EVT and 1 after SRS.

All the patients underwent staged multimodal treatment (Table 2). EVT was performed in 26 sessions, and SRS was delivered in 8 sessions, for a total of 14 shunt lesions. More than half of the patients (6/9, 67%) required 3 or more EVT sessions. Multimodal treatment combining embolization and SRS was performed in a staged and individualized manner, depending on the distribution of the shunt, hemodynamic significance, and the technical accessibility to the target compartments. Treatment-related complications occurred in 4 patients and including transient alopecia in 3 patients and intraprocedural bradycardia in 1 patient. No complications caused permanent neurological deficits.

Table 2

Treatment Summary and Long-Term Outcomes

Case No.	Target lesion	IVR sessions	Treatment methods for each lesion	Occlusion Status	Complications	Interval from last treatment to de novo dAVF development (month)	Follow-up period since the last treatment (month)	Final mRS
IVR: interventional radiology; JB: jugular bulb; Lt.: left; mRS: modified Rankin Scale; NBCA: N-butyl cyanoacrylate; ONYX: ethylene-vinyl alcohol copolymer; PTA: percutaneous transluminal angioplasty; Rt.: right; SRS: stereotactic radiosurgery; SSS: superior sagittal sinus; TAE: transarterial embolization; TS: transverse sinus; TSS: transverse sigmoid sinus; TVE: transvenous embolization *de novo.
1	1) Lt. TSS	2	1) TVE, TAE (NBCA & ONYX)	1) Partial	−	−	67	0
	2) Confluence		2) TAE (ONYX) → SRS	2) Partial
	3) SSS		3) TAE (ONYX)	3) Partial
	4) Lt. JB		4) SRS	4) Complete
2	1) Lt. TSS	2	1) TAE (NBCA) → recurrence: SRS	1) Complete	−	−	20	1
	2) Confluence		2) SRS → recurrence: TAE (ONYX)	2) Complete
3	1) Lt. TSS	3	1) TAE (ONYX)	1) Complete	−	−	25	1
	2) Confluence		2) TAE (ONYX)	2) Complete
	3) SSS		3) TAE (ONYX) ×2	3) Complete
	4) Rt. TSS (stenosis)		4) sinus PTA
4	1) Lt. TSS	3	1) TVE → TAE (ONYX)	1) Complete	alopecia	−	67	0
	2) Lt. JB		2) TVE	2) Complete
5	1) SSS	2	1) TAE (ONYX) ×2	1) Partial	alopecia	−	37	0
	2) Lt. TSS		2) SRS	2) Partial
	3) Confluence		3) SRS	3) Partial
6	1) SSS	4	1) TAE (NBCA) → TAE (ONYX)	1) Complete	alopecia	14	48	0
	2) Lt. TSS*		2) TAE (NBCA & ONYX)	2) Complete
	3) Confluence*		→ recurrence: TAE (ONYX) → SRS	3) Complete
			3) SRS
7	1) Rt. TSS	4	1) TVE & TAE (NBCA) → SRS	1) Partial	−	31	67	2
	2) Confluence		2) TVE & TAE (NBCA) ×2 → SRS	2) Partial
	3) Lt. TS*		3) TVE & TAE (NBCA) → SRS	3) Complete
8	1) Confluence	3	1) TAE (ONYX)	1) Complete	bradycardia	3	42	1
	2) Rt. TSS		2) TAE (ONYX) & TVE	2) Complete
	3) SSS*		3) TAE (ONYX) → SRS	3) Complete
9	1) SSS	3	1) TAE (NBCA) ×2	1) Complete	−	3	70	0
	2) Lt. TSS		2) TAE (NBCA) → TAE (NBCA) → SRS	2) Complete
	3) Confluence		3) TAE (NBCA) → SRS	3) Complete

The mean follow-up period after the final treatment was 49 months (range: 20-70 months). At the last follow-up, all patients showed favorable clinical outcomes, with a modified Rankin Scale (mRS) score of 0–2.

Illustrative case 1

A patient in their 70s presented with progressive pulsatile tinnitus and cognitive decrease. MRI and DSA revealed a diffuse intracranial dAVF with shunt involvement centered on the right TSS and torcular confluence (Figure 1A). Two-stage EVT comprising TAE with Onyx and TVE with coils achieved complete occlusion of the initially identified right TSS and torcular confluence shunt (Figure 1B). However, follow-up MRI and DSA performed 3 months later revealed newly detected shunt lesion in the SSS, which was not identifiable in the initial angiographic study (Figure 1C1, C2). TAE with Onyx was performed for the newly detected SSS shunt, causing a partial reduction in shunt flow. Subsequently, Gamma Knife SRS was performed on the residual SSS shunt (Figure 1D1, D2). Three months after the SRS, the irradiated shunt was completely occluded. At 38 months after the final treatment, follow-up imaging indicated sustained occlusion of all treated shunts, without further shunt detection. A favorable neurological status was maintained in the patient (Figure 1E).

Figure 1

Illustrative case of a diffuse intracranial dural arteriovenous fistula involving interconnected dural venous sinus compartments. (A) Initial angiography in a patient presenting with pulsatile tinnitus and cognitive decrease, indicating shunt lesion centered on the right transverse sigmoid sinus (TSS) and the torcular confluence. (B) Complete occlusion of the right TSS and torcular confluence shunts after transarterial embolization (TAE) with Onyx and transvenous embolization (TVE) with coils. (C1, C2) At 3 months after endovascular treatment, follow-up imaging shows newly detected shunt lesion in the superior sagittal sinus (SSS) (arrow), which had not been discernible on the initial angiographic study. (D1) Partial reduction of the newly detected SSS shunt after TAE with Onyx. (D2) Gamma Knife radiosurgery for the residual SSS shunt. (E) Complete obliteration of the irradiated SSS shunt 3 months after Gamma Knife radiosurgery. Follow-up imaging obtained 38 months after the final treatment showing persistent disappearance of all treated shunts without further interval shunt detection.

Illustrative case 2

A patient in their 20s presented with seizures. Initial angiography revealed a diffuse intracranial dAVF centered on the SSS (Figure 2A1) accompanied by PPP, indicating marked venous congestion. Initial TAE with NBCA was performed (Figure 2A2), followed by staged TAE with Onyx, which yielded complete occlusion of the SSS shunt (Figure 2B1, B2). However, follow-up imaging performed 15 months later showed newly detected shunt lesion in the left TSS (Figure 2C). TAE with Onyx was subsequently performed on this newly affected lesion, and complete occlusion was achieved (Figure 2D1, D2). Six months later, follow-up imaging revealed recurrent shunt flow in the left TSS (Figure 2E1, E2). Therefore, additional TAE with Onyx, followed by Gamma Knife SRS, was performed, causing the disappearance of the residual shunt; 45 months after the final treatment, follow-up imaging indicated sustained disappearance of all treated shunts without further interval shunt detection (Figure 2F).

Figure 2

Illustrative case of diffuse intracranial dural arteriovenous fistula showing newly detected shunt involvement after treatment and subsequent local recurrence. (A1) Initial angiography in a patient presenting with seizures, indicating shunt involvement centered on the superior sagittal sinus (SSS) with a dilated cortical vein. (A2) Angiography obtained after initial transarterial embolization (TAE) with N-butyl cyanoacrylate (NBCA), showing partial reduction of the SSS shunt. (B1, B2) Complete occlusion of the SSS shunt after staged transarterial embolization (TAE) with Onyx. (C) Follow-up angiography acquired 15 months later showing newly detected shunt involvement in the left transverse-sigmoid sinus (TSS) (arrow). (D1, D2) Complete occlusion of the newly involved left TSS after TAE with Onyx. (E1, E2) Follow-up imaging obtained 6 months later indicating recurrent shunt flow in the left TSS. (F) Follow-up imaging obtained 45 months after additional TAE with Onyx followed by Gamma Knife stereotactic radiosurgery, indicating sustained disappearance of all treated shunts without further interval shunt detection.

Discussion

Previous reports on multiple dAVFs have included heterogeneous entities ranging from lesions arising at clearly separate sites to more extensive diseases distributed across adjacent dural sinus compartments.^3,4,8) In this consecutive, single-center cohort, 8 of the 9 patients had shunt involvement including the torcular confluence, and shunting across the TSS, torcular confluence, SSS, and JB often appeared within a continuous venous network, suggesting compartmental extension or interval evolution of a related pathological process rather than the simple coexistence of several isolated fistulas. These diffuse intracranial dAVFs involving interconnected dural sinus compartments showed complex multi-compartment shunt lesion at presentation and continued angiographic changes during follow-up, with newly detected shunt lesion in another dural venous sinus compartment and recurrent shunt flow after apparent occlusion. Nevertheless, favorable long-term functional outcomes were maintained using an individualized, staged, and multimodal treatment strategy. These findings suggest that this disease subset is characterized by a dynamic temporal progression and should be managed with particular attention to venous hemodynamics and long-term follow-up.

From a therapeutic perspective, our experience supports a staged and multimodal strategy tailored to each individual patient. EVT was the first-line treatment for symptomatic or hemodynamically significant compartments, whereas SRS was performed only in case of residual low-flow shunts, lesions not amenable to safe or complete EVT, and specifically the torcular confluence lesions for which preservation of physiological venous drainage was considered important. Previous case reports and small series have described individualized combinations of embolization, sinus reconstruction, and adjunctive radiosurgery for complex intracranial dAVFs with multicompartment involvement.^6,9,10) In this study, sinus angioplasty in 1 patient improved venous outflow, which is consistent with recent reports suggesting that recanalization of impaired venous sinuses can contribute to hemodynamic improvement in certain cases.⁹⁾ Radiosurgery was not the central treatment modality in this series; however, it served as a useful adjuvant therapy for carefully selected residual low-flow shunts. Overall, these findings suggest that in this disease subset, hemodynamic control and preservation of the favorable neurological condition may be more meaningful therapeutic goals than the immediate angiographic eradication of every detectable shunt.

Another important finding in this series was that angiographic progression was not limited to the early post-treatment period. Ha et al.³⁾ reported that additional lesions in multiple metachronous dAVFs were detected 12-92 months after treatment. Subsequent reviews have likewise emphasized the potential roles of venous hypertension, sinus pathology, and long-term angiographic evolution in this disease spectrum.^3,4) Recent studies have documented newly developed dAVFs after EVT.⁵⁾ In parallel, the CONDOR study revealed that recurrence can occur even after angiographically confirmed cure, with a 3-year cumulative recurrence rate of 11% in cranial dAVFs.¹¹⁾ In our case series, newly detected shunt lesions were identified 3-31 months after treatment, and recurrent shunt flow was observed after complete occlusion. These results support the need for systematic follow-up, even after favorable early angiographic results. In our practice, MRI is performed at 3 months and every 6 months thereafter, with DSA reserved for suspected recurrence or newly detected shunt lesion. Such long-term follow-ups may allow clinically relevant interval changes to be recognized and treated before irreversible neurological deterioration occurs. On the basis of these observations, surveillance should begin in the early post-treatment period and continue for at least 5 years in patients with diffuse intracranial dAVFs involving the interconnected dural venous sinus compartments.

This study had several limitations. First, this was a retrospective, single-center study with a small number of patients, which limited the generalizability of the findings. Second, the treatment selection was based on institutional practice and physician judgment; therefore, the external validity of our staged strategy may have been limited. Third, although the mean follow-up period was 49 months, longer observation periods may be necessary to fully assess late-stage angiographic progression in this disease group. Finally, because this study was reinterpreted as having diffuse disease involving the interconnected dural venous sinus compartments, a direct comparison with earlier reports of classical multiple dAVFs should be drawn with caution.

In conclusion, diffuse intracranial dAVFs involving interconnected dural venous sinus compartments may follow an evolving clinical and angiographic course with interval progression of shunting and recurrence during follow-up. Although complete angiographic occlusion of all shunts is not always possible, favorable long-term clinical outcomes may be achieved with an individualized, staged, multimodal strategy focused on hemodynamic stabilization and long-term follow-up.

Conflicts of Interest Disclosure

All authors have no conflict of interest.

Ethics Statement

The institutional review board (IRB) of Miyazaki University Hospital approved this study (IRB number O-1881).

Informed Consent

Consent for publication was not required because the submission did not include any images or information that may identify persons.

References

• 1)  Awad IA, Little JR, Akrawi WP, et al. Intracranial dural arteriovenous malformations: factors predisposing to an aggressive neurological course. J Neurosurg. 1990;72 (6):839-50. doi: 10.3171/jns.1990.72.6.0839
• 2)  Cognard C, Gobin YP, Pierot L, et al. Cerebral dural arteriovenous fistulas: clinical and angiographic correlation with a revised classification of venous drainage. Radiology. 1995;194 (3):671-80. doi: 10.1148/radiology.194.3.7862961
• 3)  Ha SY, Kwon YS, Kim BM, et al. Clinical and angiographic characteristics of multiple dural arteriovenous shunts. AJNR Am J Neuroradiol. 2012;33 (9):1691-5. doi: 10.3174/ajnr.A3054
• 4)  Guo Y, Yu J, Zhao Y, et al. Progress in research on intracranial multiple dural arteriovenous fistulas: a review. Biomed Rep. 2017;8 (1):17-25. doi: 10.3892/br.2017.1021
• 5)  Duquette E, Dowlati E, Abdullah T, et al. De novo dural arteriovenous fistulas after endovascular treatment: case illustration and literature review. Interv Neuroradiol. 2024;30 (4):597-603. doi: 10.1177/15910199221118517
• 6)  Chen CC, Cho YD, Yoo DH, et al. Endovascular management of multiple intracranial dural arteriovenous fistulas. J Neuroradiol. 2019;46 (6):390-7. doi: 10.1016/j.neurad.2018.10.010
• 7)  Brinjikji W, Cloft HJ, Lanzino G. Clinical presentation and imaging findings of patients with dural arteriovenous fistulas with an angiographic pseudophlebitic pattern. AJNR Am J Neuroradiol. 2020;41 (12):2285-91. doi: 10.3174/ajnr.A6811
• 8)  Fiumara E, Tumbiolo S, Bellomonte ML, et al. Resection of the transverse sinuses and confluence of sinuses for treatment of multiple dural arteriovenous fistulas. Case report. J Neurosurg. 2004;100 (2):348-52. doi: 10.3171/jns.2004.100.2.0348
• 9)  Zhao TY, Xu GQ, Chen ZC, et al. Dural arteriovenous fistula may be occluded through recanalization of impaired venous sinuses. Medicine (Baltimore). 2024;103 (41):e40097. doi: 10.1097/MD.0000000000040097
• 10)  Su X, Song Z, Ma Y, et al. Multiple synchronous intracranial dural arteriovenous fistulas. Neurosurg Rev. 2025;48 (1):535. doi: 10.1007/s10143-025-03696-7
• 11)  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):981-9. doi: 10.3171/2021.1.JNS202033