Intracranial Dural Arteriovenous Fistulas with Deep Venous Drainage: A Single-Center Retrospective Cohort Study

doi:10.21203/rs.3.rs-5242689/v1

Objective

Dural arteriovenous fistulas (DAVFs) with deep venous drainage (DVD) (DAVFs-DVD) are characteristically associated with nonhemorrhagic neurological deficits, most notably cognitive impairment. Large studies have yet to thoroughly characterize these DAVFs. We conducted an analysis of the largest cohort of DAVFs-DVD to provide a comprehensive characterization of this specific subset.

Methods

This study was a retrospective cohort analysis of 207 patients with DAVFs-DVD who were hospitalized at a single medical center between 2002 and 2022. Logistic regression was employed to identify angiographic features specific to DAVFs associated with cognitive impairment. Additionally, post-treatment outcomes were evaluated.

Results

DAVFs-DVDs can be classified into two categories: the Galenic group, where arterial feeders drain directly into the vein of Galen (VoG), and the non-Galenic group, where arterial feeders drain into other veins or venous sinuses and indirectly flow into the VoG. Cognitive impairment accounting for 15.9% of all clinical presentations. We found that the overall rate of complete occlusion was 86.6%, while the overall rate of complications was 16.4%. Moreover, patients with DAVFs-DVD presenting with cognitive impairment were more likely to have straight sinus stenosis or occlusion, as well as DAVFs located at the midline.

Conclusion

DAVFs-DVDs can be classified into two categories: the Galenic group and the non-Galenic group. Patients with DAVFs-DVD presenting with cognitive impairment were more likely to have straight sinus stenosis or occlusion, as well as DAVFs located at the midline. Venous hypertension of the deep venous system is a key angiographic feature of DAVFs associated with thalamic cognitive impairment.

Dural arteriovenous fistula

Dementia

Angiography

Outcomes

Intracranial dural arteriovenous fistulas (DAVFs) are abnormal connections between dural or pial arteries and the dural venous sinuses or cortical veins. They represent approximately 10–15% of all intracranial vascular malformations.^1,2 Patients with DAVFs can present with benign symptoms associated with increased sinus drainage, such as tinnitus and ophthalmologic symptoms, or with more severe symptoms linked to critical cortical venous hypertension, including intracranial hemorrhage and non-hemorrhagic neurological deficits.³

Cognitive impairment caused by DAVFs can be categorized into two types. One is due to thalamic involvement, which is often seen in DAVFs with deep venous drainage (DVD) (DAVFs-DVD).^4–8 The deep venous system (DVS) of the brain is composed of the internal cerebral veins (ICVs), basal veins, great veins (the vein of Galen, the VoG), and their respective tributaries. The ICVs collect numerous small tributaries from structures such as the fornix, hippocampal commissure, choroid plexus of the third ventricle, and the thalamus.⁹ Retrograde flow into the DVS can lead to venous congestion in the bilateral thalami and basal ganglia, resulting in cognitive impairment. The other type of cognitive impairment caused by DAVFs is due to cortical venous hypertension.^4,6,8 Disruption of the superficial cortical venous drainage system and subcortical white matter can lead to venous hypertension, accompanied by congestion in the medullary veins. Patients with this drainage pattern often exhibit symptoms of cortical dysfunction, including decreased attention span, acalculia, and executive impairment.¹⁰ Most patients with DAVFs who present with cognitive impairment experience a combination of cortical and thalamic venous congestion.⁴

Recently, a large-sample study used propensity score matching analysis to provide a detailed characterization of DAVFs that lead to cognitive impairment.⁴ They found that venous hypertension is a key angiographic feature of DAVFs associated with cognitive impairment. Additionally, these fistulas exhibit a complex angioarchitecture, characterized by an increased number of arteriovenous connections and stenotic sinuses. The presence of venous ectasia further worsens drainage impairment, contributing to the development of cognitive impairment in patients with DAVFs. However, the description of DAVFs-DVD is largely limited to case series, with few comprehensive studies available.^11,12 In this study, we present the angiographic determinants and postprocedural outcomes of the largest cohort of DAVFs-DVD to date.

Study design

This study was a retrospective cohort analysis of 207 patients with DAVFs-DVD who were hospitalized at a single medical center between 2002 and 2022. The current investigation rigorously conformed to the Strengthening the Reporting of Cohort, Cross-Sectional, and Case–Control Studies in Surgery 2021 (STROCSS 2021,Supplemental Digital Content 1, http://links.lww.com/JS9/B513) guidelines, as delineated for observational research endeavors.¹³ This retrospective study was registered on ClinicalTrials.gov (NCT06543472). Given the retrospective nature of the study, the requirement for informed consent was appropriately waived. Relevant data were obtained from digital medical records, with all personally identifiable information, including names, addresses, and social security numbers, removed and replaced by unique codes to ensure that the data could not be traced back to individual patients.

Data Acquisition and Analysis

The collected data included patient demographics, clinical presentation, DAVF location, feeding arteries, drainage patterns, Borden/Cognard classification, treatment modalities, angiographic outcomes, procedure-related complications, and follow-up information. Flow related symptoms were defined as symptoms related to increased venous drainage, including tinnitus, bruits, chemosis/proptosis, ophthalmoplegia/diplopia, and isolated headaches not attributable to hemorrhage. Non-hemorrhagic neurological deficits (NHNDs) were defined as focal or global neurological deficits resulting from venous hypertension. NHNDs encompassed cognitive impairment, seizures, ataxia, sensory and motor deficits, aphasia/dysarthria, cranial nerve palsies (excluding cranial nerves III, IV, and VI), and hydrocephalus. Cognitive impairment included recently diagnosed symptoms or a sudden worsening of subacute symptoms, leading to impaired cognitive function, psychiatric disturbances, impaired language production, and episodes of altered consciousness. The presence of a congestive pseudophlebitic appearance was noted as cortical venous strain.¹⁴ Various angiographic characteristics were collected, including the number of arterial feeders, the number of draining veins, and the presence of sinus stenosis/occlusion or sinus/venous ectasia. DAVFs-DVDs are classified into two categories: the Galenic group, where arterial feeders drain directly into the VoG, and the non-Galenic group, where arterial feeders drain into other veins or venous sinuses and indirectly flow into the VoG (Fig. 1). Since all the included cases involved DAVFs-DVD, the resulting cognitive impairment can be classified as the thalamic variant of cognitive impairment.^4,8

Follow-up and outcomes

All forms of patient monitoring, including admissions, outpatient visits, and telephone follow-ups, were considered. Radiological follow-up was limited to digital subtraction angiography (DSA) imaging. Follow-up duration was defined as the period from the last treatment to the most recent DAVF-related follow-up. The primary outcome was determined by the modified Rankin Scale (mRS) score at the final clinical follow-up.

Clinical data and angiographic images were reviewed by three neurosurgeons (X.S., X.Y.L., and Z.H.S.). Additionally, three experienced neurosurgeons (Y.J.M., H.Q.Z., and P.Z.) independently verified the embolization outcomes, angioarchitecture, and DSA follow-up imaging. Any disagreements were resolved through discussion.

Statistical analysis

All statistical analyses were conducted using R (version 4.2.3). Baseline characteristics were compared between the Galenic and Non-Galenic group. Continuous variables were compared using Student’s t-test or Wilcoxon rank-sum tests, while categorical variables were compared using Pearson’s χ² or Fisher’s exact tests, as appropriate. Univariate analysis was performed to evaluate covariates predictive of cognitive impairment. Factors identified as significant in the univariate analysis (P < 0.05) were subsequently included in multivariate logistic regression models for cognitive impairment. Statistical significance was set at P < 0.05, and all tests were two-tailed. Missing data were not imputed.

A total of 207 patients with DAVFs-DVD are included in this study. The mean age of patients with DAVFs-DVD was 49.2 ± 12.3 years, and 69.1% (143/207) were men (Table 1).

Table 1

Demographics and presenting characteristics
	Total (n = 207)	Galenic group (n = 43)	Non-Galenic group (n = 164)	P value
Mean age, yrs	49.2 ± 12.3	48.3 ± 10.7	49.4 ± 12.6	0.435
Sex				0.007
Male	143 (69.1)	37 (86.1)	105 (64.6)
Female	64 (30.9)	6 (14.0)	58 (35.4)
Presentation				0.877
Asymptomatic/Symptoms, unrelated to DAVF	15 (7.3)	4 (9.3)	11 (6.7)
Hemorrhage	36 (17.4)	7 (16.3)	29 (17.7)
NHND	71 (34.3)	16 (37.2)	55 (33.5)
Flow-related	85 (41.1)	16 (37.2)	69 (42.1)
Cognitive decline	33 (15.9)	13 (30.2)	20 (12.2)	0.004
Location
Galenic	43 (20.8)	43 (100)	0
Transverse/sigmoid sinus	46 (22.2)	0	46 (28.0)
Other tentorial	86 (41.5)	0	86 (52.4)
Cavernous sinus	13 (6.3)	0	13 (7.9)
Jugular foramen	2 (1.0)	0	2 (1.2)
Sylvian/middle cranial fossa	3 (1.4)	0	3 (1.8)
Convexity/superior sagittal sinus	13 (6.3)	0	13 (7.9)
Anterior cranial fossa	1 (0.5)	0	1 (0.6)
mRS score at presentation				0.682
Mean	1.96 ± 1.12	1.88 ± 1.18	1.98 ± 1.11
0	7 (3.4)	3 (7.0)	4 (2.4)
1	75 (36.2)	15 (34.9)	60 (36.6)
2	77 (37.2)	16 (37.2)	61 (37.2)
3	22 (10.6)	3 (7.0)	19 (11.6)
4	20 (9.7)	5 (11.6)	15 (9.2)
5	6 (2.9)	1 (2.3)	5 (3.1)
Values represent the number of patients/total number of patients with data (%) or mean ± SD unless stated otherwise. Boldface type indicates statistical significance.
NHND, Non-hemorrhagic neurological deficit.

Presentation and angioarchitecture

36.2% (75/207) of DAVFs-DVD were Borden type Ⅱ, and 63.8% (132/207) were Borden type Ⅲ. 20.8% (43/207) of DAVFs exhibited a direct arteriovenous shunt involving the VoG (Galenic group), while the remaining DAVFs drained into the VoG indirectly (Non-Galenic DAVFs). The most common location of DAVFs in the non-Galenic group was the tentorial region (52.4%), followed by the transverse/sigmoid sinus region (28.0%). Compared to patients with non-Galenic DAVFs, those with Galenic DAVFs show a higher incidence of cognitive decline and a greater proportion of male patients (Table 1).

Galenic DAVFs had a greater number of pial arterial feeders (P < 0.001), a higher proportion of venous ectasia (P = 0.006), and higher Borden/Cognard grades (P < 0.001) compared to non-Galenic DAVFs (Table 2). Venous sinus obstruction was observed in 48.8% of DAVFs-DVD. In six cases (14.0%) of Galenic DAVFs, stenosis or occlusion of the straight sinus (SS) was observed, directly causing retrograde flow into the ICVs. Among the 207 cases of DAVFs-DVD, 15.9% (33/207) of patients exhibited cognitive impairment, and 30.3% (10/33) of those with venous drainage showed SS stenosis or occlusion. Among the 13 Galenic DAVF patients who exhibited cognitive decline, seven (53.8%) had obstructed drainage veins, and 8 (61.5%) experienced direct retrograde flow into the ICVs. Among the 20 non-Galenic DAVF patients who exhibited cognitive decline, 14 (70%) had obstructed drainage veins, and 11 (55%) experienced direct retrograde flow into the ICVs. A multivariate logistic regression model revealed that DAVFs-DVD patients presenting with cognitive impairment were more likely to have a higher number of SS stenoses or occlusions (OR 8.76; 95% Cl 1.87–41.14, P = 0.006), as well as DAVFs located at the midline (OR 3.36; 95% Cl 1.34–8.42, P = 0.01) (Table 5).

Table 2

Angioarchitecture and radiographic grade characteristics
	Total (n = 207)	Galenic group (n = 43)	Non-Galenic group (n = 164)	P value
Pial artery supply	100 (48.3)	38 (88.4)	62 (37.8)	<0.001
Spinal medullary venous drainage	9 (4.4)	0 (0)	9 (5.5)	0.731
Venous ectasia	116 (56.0)	32 (74.4)	84 (51.2)	0.006
Cortical venous strain	88 (42.5)	15 (34.88)	73 (44.5)	0.256
Venous sinus outflow
Stenosis	18 (8.7)	2 (4.7)	16 (9.8)	0.451
Occlusion	83 (40.1)	18 (41.9)	65 (39.6)	0.791
Borden type				<0.001
Ⅱ	75 (36.2)	0 (0)	75 (45.7)
Ⅲ	132 (63.8)	43 (100)	89 (54.3)
Cognard type				<0.001
Ⅱb	5 (2.4)	0 (0)	5 (3.1)
Ⅱa + b	70 (33.8)	0 (0)	70 (42.7)
Ⅲ	37 (17.9)	11 (25.6)	26 (15.9)
Ⅳ	85 (41.1)	32 (74.4)	53 (32.3)
Ⅴ	10 (4.8)	0 (0)	10 (6.1)
Values represent the number of patients/total number of patients with data (%) or mean ± SD unless stated otherwise. Boldface type indicates statistical significance.

Table 3. Summary of treatment modalities and success rate

	Total (n=207)	Galenic group (n=43)	Non-Galenic group (n=164)	P value
Endovascular embolization	196/207 (94.7)	40/43 (93.0)	156/164 (95.1)	0.870
No. of embolizations	2.07±2.08 (1-18)	2.53±1.84 (1-10)	1.96±2.12 (1-18)	＜0.001
Mean (range)	2.07±2.08 (1-18)	2.53±1.84 (1-10)	1.96±2.12 (1-18)
1	116 (59.2)	14/40 (35.0)	102/156 (65.4)
2	37 (18.9)	13/40 (32.5)	24/156 (15.4)
≥3	43 (21.9)	13/40 (32.5)	30/156 (19.2)
Complete obliteration	161/196 (82.1)	32/40 (80.0)	129/156 (82.7)	0.692
Embolization approach				0.006
Trans-arterial	159/196 (81.1)	39/40 (97.5)	120/156 (76.9)
Transvenous	18/196	1/40 (2.5)	18/156 (11.5)
Both	19/196	0/40 (0)	18/156 (11.5)
Surgery	13/207 (6.3)	0/43 (0)	13/164 (7.9)	0.120
Complete obliteration			13/13 (100)
Radiosurgery	2/207 (1.0)	1/43 (2.3)	1/164 (0.6)	0.373
Conservative	6/207 (2.9)	3/43 (7.0)	3/164 (1.8)	0.200

Values represent the number of patients/total number of patients with data (%) or mean ± SD unless stated otherwise. Boldface type indicates statistical significance.

Table 4. Follow-up and clinical outcomes

	Total (n=207)	Galenic group (n=43)	Non-Galenic group (n=164)	P value
Mean total angiographic FU from last treatment, mos	91/207 (44.0)	20/43 (46.5)	71/164 (43.3)
	17.04 ± 18.01	17.90 ± 20.94	16.79 ± 17.08	0.510
Recurrence	3/81 (3.7)	0/19 (0)	3/62 (4.8)	1.000
Total clinical FU from last treatment, mos	173/207 (83.6)	36/43 (83.7)	137/164 (83.5)
	45.01 ± 47.84	44.59 ± 54.41	45.12 ± 45.96	0.945
Clinical course				0.454
Asymptomatic	91/173 (52.6)	16/36 (44.4)	75/137 (54.7)
Improved symptoms	37/173 (21.4)	8/36 (22.2)	29/137 (21.2)
Symptomatic, stable	21/173 (12.1)	4/36 (11.1)	17/137 (12.4)
Symptomatic, worsened/new	24/173 (13.9)	8/36 (22.2)	16/137 (11.7)
Death related to DAVF	10/173 (5.8)	2/36 (5.6)	8/137 (5.84)
mRS score at FU				0.755
Mean	1.26 ± 1.84	1.58 ± 2.01	1.18 ± 1.78
0	90/173 (52.0)	16/36 (44.4)	74/137 (54.0)
1	34/173 (19.7)	7/36 (19.4)	27/137 (19.7)
2	17/173 (9.8)	4/36 (11.1)	13/137 (9.5)
3	9/173 (5.2)	3/36 (8.3)	6/137 (4.4)
4	6/173 (3.5)	1/36 (2.8)	5/137 (3.7)
5	3/173 (1.7)	1/36 (2.8)	2/137 (1.5)
6	14/173 (8.1)	4/36 (11.1)	10/137 (7.3)

FU, follow-up.

Values represent the number of patients/total number of patients with data (%) or mean ± SD unless stated otherwise. Boldface type indicates statistical significance.

Table 5. Logistic regression analysis of angiographic characteristics identifying DAVF with deep venous drainage associated with cognitive impairment

	Univariate logistic regression				Multivariate logistic regression
	Variable	OR	95% Cl	P value	OR	95% Cl	P value
Baseline characteristics	Female	0.55	0.23-1.35	0.193
	Alcohol use	1.46	0.50-4.22	0.488
	Diabetes mellitus	0.53	0.12-2.38	0.404
	Hypertension	1.42	0.65-3.10	0.382
	History of venous sinus thrombosis	2.53	0.73-8.76	0.143
Angioarchitecture	Venous ectasia	0.45	0.21-0.96	0.039	0.59	0.24-1.44	0.247
	Number of arterial feeders	1.44	0.68-3.03	0.342
	Number of draining veins	0.50	0.18-1.39	0.183
	Cortical venous strain	1.54	0.73-3.25	0.256
	Midline location	5.87	2.60-13.22	＜0.001	3.36	1.34-8.42	0.010
DAVF drianage	Sinus stenosis	2.21	0.73-6.69	0.160
	Sinus occlusion	1.74	0.82-3.67	0.147
	Straight sinus stenosis or occlusion	28.50	7.38-110.11	0.006	8.76	1.87-41.14	0.006
	Internal cerebral vein	5.20	2.38-11.37	＜0.001	2.38	0.94-5.98	0.066
	The basal vein of Rosenthal	0.56	0.24-1.32	0.188

Boldface type indicates statistical significance.

Treatment outcome and follow-up

97.1% (201/207) of patients with DAVFs-DVD received treatment. Complete obliteration of the DAVFs-DVD was achieved in 86.6% (174/201) of cases. Compared to non-Galenic DAVFs, Galenic DAVFs are more commonly treated via trans-arterial embolization (TAE) (P < 0.001) and tend to require multiple embolization procedures (P = 0.006) (Table 3). A total of 33 patients with DAVFs-DVD developed complications following treatment. Of these, 28 were associated with TAE, four with transvenous embolization, and one with surgery. Of these, 11 patients experienced worsening symptoms due to treatment-related complications, and four patients died as a result—three from intracranial hemorrhage during TAE, and one from intracranial hemorrhage following surgery.

Clinical follow-up was available for 83.6% (173/207) of patients with a DAVF-DVD (Table 4). A total of 91 patients (44%) with DAVFs-DVD underwent follow-up DSA evaluations. This included 84 patients with complete fistula closure, two patients with partial fistula obliteration, one untreated patient, three patients with fistula recurrence and one patient with a newly developed DAVF. Among the DAVFs-DVD cases presenting with cognitive impairment, 84.8% (28/33) of patients underwent clinical follow-up, and 36.4% (12/33) underwent DSA follow-up. This included five asymptomatic patients, 21 patients with either symptom improvement or stable symptoms, and two patients with worsened symptoms. All 12 patients with DAVFs who underwent follow-up angiography demonstrated complete occlusion. Of these, one patient experienced worsening symptoms, four remained asymptomatic, and the others showed either symptom improvement or stable. A total of 8.1% (14/173) of patients died during clinical follow-up, ten of which were related to DAVF and its treatment.

In this study, we present an analysis of the largest cohort of patients with DAVFs-DVD. DAVFs-DVDs can be classified into two categories: the Galenic group, where arterial feeders drain directly into the VoG, and the non-Galenic group, where arterial feeders drain into other veins or venous sinuses and indirectly flow into the VoG. Cognitive decline is a characteristic clinical manifestation of DAVFs-DVD, accounting for 15.9% of all clinical presentations. We found that the overall rate of complete occlusion was 86.6%, while the overall rate of complications was 16.4% (33/201). Moreover, patients with DAVFs-DVD presenting with cognitive impairment were more likely to have SS stenosis or occlusion, as well as DAVFs located at the midline. This means that the closer the fistula was to the DVS and the more obstructed the draining veins near DVS, the higher the proportion of cases with cognitive impairment. Venous hypertension of the DVS is a key angiographic feature of DAVFs associated with thalamic cognitive impairment.

Anatomical, clinical, and angioarchitectural characteristics of DAVFs-DVD.

The venous drainage of the thalamus primarily occurs via the basal veins of Rosenthal and the ICVs, which drain into the VoG and the SS. Retrograde flow in these pathways can result in venous congestion in the bilateral thalami and basal ganglia, leading to impaired cognitive function.^4,6 In our study of 207 cases DAVFs-DVD, only 15.9% of patients exhibited cognitive impairment, and 30.3% of those with venous drainage showed SS stenosis or occlusion. In patients with DAVFs-DVD presenting with benign clinical features, including asymptomatic cases and flow-related symptoms, the SS was largely unobstructed (98%). In the study by Sanchez et al., they concluded that venous hypertension is a key angiographic feature of DAVFs associated with cognitive impairment.⁴ In our study, patients with DAVFs-DVD presenting with cognitive impairment were more likely to have SS stenosis or occlusion, as well as DAVFs located at the midline. Additionally, cognitive impairment was more prevalent in Galenic DAVFs compared to non-Galenic DAVFs. This suggests that venous hypertension of the DVS is a key angiographic feature of DAVFs associated with thalamic cognitive impairment. This may explain why DAVFs with cognitive impairment tend to have multiple arterial feeders and a higher number of draining veins.^4,8 Chronic venous hypertension likely stimulates neoangiogenesis, resulting in the formation of additional arterial feeders, shunts, and draining veins.^4,6 In our study, DAVFs with pial artery supply accounted for 48.3% of all DAVFs-DVD cases, which is significantly higher than the 10–20% typically reported for DAVFs with pial artery involvement.¹⁵ Studies have shown that venous hypertension is an independent predictive factor for the presence of pial arterial supply.^16,17 The higher proportion of pial arterial supply may suggest that DAVFs-DVD, compared to DAVFs without DVS drainage, are associated with greater venous pressure.

Treatment outcome

In this study, the overall rate of complete occlusion was 86.6%, the complication rate was 16.4%, and the recurrence/new development rate was 4.4%. Compared to a recent meta-analysis on DAVFs involving the VoG, the rates of complete occlusion and complications in our study were similar.¹² Compared to our earlier studies on DAVFs in other locations, the final complete occlusion rate for DAVFs in this region was slightly lower, while the complication rate was slightly higher. This may be closely related to the more complex angioarchitecture and the higher proportion of pial arterial supply in DAVFs-DVD. Among the DAVFs-DVD patients with cognitive impairment in our clinical follow-up, only five became completely asymptomatic after treatment, while the others experienced persistent, varying degrees of irreversible cognitive impairment. Persistent deficits may remain in some patients due to ischemia if treatment is not administered promptly.^18,19

Study Limitations

This study's retrospective nature and design introduced unavoidable selection bias and technical variability. Nevertheless, the large cohort and diverse patient population provided valuable insights into the real-world experience of DAVFs-DVD. It is also worth noting that cognitive impairment was not assessed using a standardized test battery; similar to previous studies, it was determined based on clinical judgment.^4,6,8 Furthermore, due to the study’s time span, not all patients were followed up, which may have led to an underestimation of the final complete occlusion and reucurrence rates. Lastly, as our hospital is a tertiary referral center for cerebrovascular disorders, the severity and complexity of cases were likely higher due to referral bias.

DAVFs-DVDs can be classified into two categories: the Galenic group and the non-Galenic group. Patients with DAVFs-DVD presenting with cognitive impairment were more likely to have SS stenosis or occlusion, as well as DAVFs located at the midline. Venous hypertension of the DVS is a key angiographic feature of DAVFs associated with thalamic cognitive impairment.

DAVF, dural arteriovenous fistula; DSA, digital subtraction angiography; DVD, deep venous drainage; DVS, deep venous system; ICV, internal cerebral vein; mRS, modified Ranking score; NHNDs, non-hemorrhagic neurological deficits; SS, straight sinus; TAE, trans-arterial embolization; VoG, the vein of Galen;

Ethical Approval

This retrospective study involved human participants and was approved by Ethics Committee of Xuanwu Hospital, Capital Medical University [2017; 010].

Author Contribution

Conception and design: Xin Su, Peng Zhang, Yongjie Ma, Hongqi Zhang, Zihao Song. Acquisition of data: Xin Su, Zihao Song, Yiguang Chen, Huiwei Liu, Huishen Pang, Mingyue Huang, Chao Zhang. Drafting the article: Xin Su. Critically revising the article: Yongjie Ma, Ming Ye, Liyong Sun, Tao Hong, Peng Zhang, Hongqi Zhang. All of the authors have read and approved the final manuscript.

Acknowledgement

None

Cognard C, Gobin YP, Pierot L et al (1995) Cerebral dural arteriovenous fistulas: clinical and angiographic correlation with a revised classification of venous drainage. Radiology 194(3):671–680
Borden JA, Wu JK, Shucart WA (1995) A proposed classification for spinal and cranial dural arteriovenous fistulous malformations and implications for treatment. J Neurosurg 82(2):166–179
Guniganti R, Giordan E, Chen CJ et al (2022) Consortium for Dural Arteriovenous Fistula Outcomes Research (CONDOR): rationale, design, and initial characterization of patient cohort. J Neurosurg 136(4):951–961
Sanchez S, Wendt L, Hayakawa M et al (2023) Dural Arteriovenous Fistulas With Cognitive Impairment: Angiographic Characteristics and Treatment Outcomes. Neurosurgery
Bernstein R, Dowd CF, Gress DR (2003) Rapidly reversible dementia. Lancet (London England) 361(9355):392
Brito A, Tsang ACO, Hilditch C, Nicholson P, Krings T, Brinjikji W (2019) Intracranial Dural Arteriovenous Fistula as a Reversible Cause of Dementia: Case Series and Literature Review. World Neurosurg 121:e543–e553
Matsumura A, Oda M, Hozuki T, Imai T, Shimohama S (2008) Dural arteriovenous fistula in a case of dementia with bithalamic MR lesions. Neurology 71(19):1553
Holekamp TF, Mollman ME, Murphy RK et al (2016) Dural arteriovenous fistula-induced thalamic dementia: report of 4 cases. J Neurosurg 124(6):1752–1765
Rhoton AL (2002) Jr. The cerebral veins. Neurosurgery 51(4 Suppl):S159–205
Fujii H, Nagano Y, Hosomi N, Matsumoto M (2014) Dural arteriovenous fistula presenting with progressive dementia and parkinsonism. BMJ case Rep ;2014
Islak C, Bagcilar O, Nacar Dogan S et al (2022) Endovascular management of anterior falcotentorial dural arteriovenous fistulas: importance of functionality of deep venous system and existence of accompanying choroidal arteriovenous malformation. J neurointerventional Surg 14(6):599–604
Lauzier DC, Ullman H, Hardi A, Derdeyn C, Cross DT, Moran CJ (2023) Endovascular treatment of dural arteriovenous fistulas involving the vein of Galen: a single-center cohort and meta-analysis. J neurointerventional Surg
Mathew G, Agha R, Albrecht J et al (2021) STROCSS. : Strengthening the reporting of cohort, cross-sectional and case-control studies in surgery. International journal of surgery (London, England). 2021;96:106165
Willinsky R, Goyal M, terBrugge K, Montanera W (1999) Tortuous, engorged pial veins in intracranial dural arteriovenous fistulas: correlations with presentation, location, and MR findings in 122 patients. AJNR Am J Neuroradiol 20(6):1031–1036
Su X, Ma Y, Song Z, Zhang P, Zhang H (2024) Intracranial dural arteriovenous fistulas with pial arterial supply: A narrative review. Brain Circulation 10(3):205–212
Osada T, Krings T (2019) Intracranial Dural Arteriovenous Fistulas with Pial Arterial Supply. Neurosurgery 84(1):104–115
Brinjikji W, Cloft HJ, Lanzino G (2021) Clinical, angiographic, and treatment characteristics of cranial dural arteriovenous fistulas with pial arterial supply. J neurointerventional Surg 13(4):331–335
Chen PM, Olson SE, Handwerker J (2020) Bithalamic Lesions: Cranial Dural Arteriovenous Fistula Manifesting as Thalamic Dementia. Stroke 51(12):e355–e358
Su X, Song Z, Tu T et al (2023) A retrospective study of 78 adult tentorial middle line region dural arteriovenous fistulae. Acta Neurol Belgica 123(4):1395–1404

No competing interests reported.

Intracranial Dural Arteriovenous Fistulas with Deep Venous Drainage: A Single-Center Retrospective Cohort Study

Status:

Version 1

Abstract

Objective

Methods

Results

Conclusion

Figures

INTRODUCTION

METHODS

Study design

Data Acquisition and Analysis

Follow-up and outcomes

Statistical analysis

Results

Presentation and angioarchitecture

Treatment outcome and follow-up

Discussion

Treatment outcome

Study Limitations

Conclusion

Abbreviations

Declarations

Author Contribution

Acknowledgement

References

Additional Declarations

Status:

Version 1