Forearm-only transarterial and transvenous approach for endovascular intervention of dural arteriovenous fistula

Abstract

Background

Transradial arterial access has become widely used as a less invasive approach in neuroendovascular therapy, but the forearm venous approach has rarely been reported.

Objectives

This study aimed to assess the safety and efficacy of forearm transvenous neurointervention for intracranial lesions. We present our experience with a simultaneous forearm-only transarterial and venous approach (fTAVA) for dural arteriovenous fistulas (dAVFs).

Methods

We retrospectively reviewed a prospective database of consecutive patients who underwent fTAVA for dAVFs between 2021 and 2024. fTAVA was performed using the right radial artery and superficial forearm vein as puncture sites. Arterial closure was achieved using a radial compression device, whereas bandage compression was used for venous closure. Procedural success, angiographic results, procedure-related complications and patient satisfaction were evaluated.

Results

Overall, 13 (8 carotid-cavernous sinus and 5 transverse-sigmoid sinus fistulas) procedures using fTAVA were successfully performed with favorable outcomes. Arterial puncture was performed at the distal radial artery in nine cases. The venous puncture site was the median cubital vein in nine cases and the forearm cephalic vein in four cases. The targeted fistulas were distal to the right jugular vein in four cases and the left jugular vein in nine cases. They were successfully accessed in all cases. The angiographic result was total occlusion in eleven cases and subtotal occlusion in two cases. There was one patient with minor access-site complication in distal radial artery.

Conclusions

The fTAVA is a safe and effective method for the endovascular treatment of dAVFs and is associated with reduced patient discomfort.

Introduction

The transfemoral arteriovenous approach is commonly employed in the endovascular treatment of dural arteriovenous fistulas (dAVFs). Recent advancements in liquid embolic materials have expanded opportunities for transarterial embolization; however, many cases still necessitate transvenous embolization. Cannulation of both vessels via the transfemoral arteriovenous approach can be challenging, even with ultrasound guidance, particularly in obese patients, and complications at the puncture site remain common, especially among the patients on antithrombotic agents. In contrast, the transupper limb venous approach is associated with a lower risk of puncture-site complications and offers comparable controllability.^1,2 Despite its advantages, reports on this technique are limited, and it has yet to be standardized for neurointerventional procedures. Abecassis et al. reviewed a large case series involving forearm transvenous access. ² However, their study, which includes various diagnostic techniques such as venous manometry, balloon occlusion, and sampling, focuses on access success rather than clinical outcomes.

Herein, we present our experience with a simultaneous forearm-only transarterial and venous approach (fTAVA) for dAVFs, emphasizing technical nuances and clinical outcomes.

Methods

Study design, enrollment, and patients’ selection

This study retrospectively analyzed a database of patients diagnosed with dAVF requiring endovascular intervention who were consecutively admitted to our institution between 2021 and February 2024. During this period, when the fTAVA was implemented as a first-line therapy for the transvenous approach, 16 patients were identified as candidates for neurointervention targeting dAVFs. Among these, 13 patients underwent fTAVA, while three were excluded due to treatment involving transarterial fistula occlusion with liquid embolic material.

The primary endpoint of this study included procedural success with fTAVA. Failure was defined as inability to obtain or use fTAVA, necessitating crossover to femoral or jugular vein. The secondary endpoint was procedure-related complications. Additionally, patient-reported discomfort after the procedure and at discharge was assessed through interviews.

This study was approved by the institutional review board (Approval no.: 5301) and was conducted in accordance with the principles of the Declaration of Helsinki. The requirement for informed consent was waived due to the retrospective design of the study.

Procedure and imaging

The procedure was performed in a digital angiography suite. Under general anesthesia, the patient was positioned with the forearm on the arm board and the forearm abducted gently to facilitate forearm venous puncture. Both the radial artery and forearm vein were punctured under ultrasound guidance. The angiographic table was rotated clockwise by approximately 15–20° to visualize the right forearm with a single-plane set-up, enabling anteroposterior fluoroscopic and roadmap imaging.

Arterial procedure

The right radial artery was used in all cases. The choice of distal or proximal radial access was determined based on the vessel diameter on the ultrasound examination. The indication for puncture was an arterial vessel diameter of ≥1.8 mm. The primary puncture site was the distal radial artery, located in the anatomical snuffbox. The proximal radial artery was selected for cases in which the diameter of the distal radial artery was <1.8 mm. Under ultrasound guidance, a 4-French sheath was successfully cannulated into the radial artery after the puncture with a 22-gauge needle using the double-wall technique. Subsequently, forearm venogram was performed to aid in cannulation of the catheter system into a suitable forearm vein.

Venous procedure

For transvenous access, the median cubital vein was primarily used as the access site because it was easily accessible and palpable. We prepared a tourniquet to increase the prominence of the veins for easier puncture. After confirming the appropriate venous route under the venogram through the arterial sheath, the targeted vein was punctured using a 22-gauge needle under ultrasound guidance, after applying a tourniquet to the upper arm. Thereafter, a 4-French guiding sheath (Fubuki, ASAHI INTECC, Japan) with a dilator was inserted directly along the guidewire. Systemic heparinization (5000 units) was performed intravenously after completion of both arterial and venous cannulation.

Approach for the targeted lesion and puncture site closure

A 4-French guiding sheath and 125-cm 4-French JBII coaxial catheter were coaxially navigated from the right superficial forearm vein using a standard 0.035-inch guidewire. During the guidance from the superficial forearm vein to the right subclavian vein, venogram through the venous catheter could ensure safe navigation, avoiding venous perforation and injury due to various venous anastomoses and steep angulation. Simultaneously, a 125-cm 4-French Simmons diagnostic catheter via the arterial sheath was positioned in the main feeding artery. To safely navigate the guiding sheath to the jugular vein, venographic mapping via a transarterial catheter placed at the feeding vessel was useful to visualize each jugular and innominate vein or the superior vena cava and subclavian vein. After reaching the jugular bulb, the targeted lesion should be accessed using an intermediate coaxial catheter and a microcatheter, replacing the JBII coaxial catheter. The intermediate coaxial catheter was essential to gain sufficient support. Coil embolization was performed after advancing the microcatheter into the targeted fistula. For closure of the puncture site, a compression device and a bandage were applied to the arterial and venous puncture sites, respectively.

Data availability

The datasets generated and/or analyzed during the current study are not publicly available due to ethical restrictions but are available from the corresponding author on reasonable request.

Results

A total of 13 procedures in 13 patients were successfully performed with coil embolization using fTAVA. The characteristics of the 13 patients are shown in Table 1. The median age was 73 years (range, 28–81 years), and there were 8 women. The median body mass index was 23.4 kg/m² (range, 21.5–35.2). The site of the dAVF was the cavernous sinus in eight cases and the transverse-sigmoid sinus in five cases, and the targeted jugular vein was the left jugular vein in nine cases. According to the Borden classification, there were five cases of type I, seven cases of type II, and one case of type III.

Table 1.

Patient characteristics (n = 13).

Case no.	Age (years)	Sex	Body mass index	The use of antithrombotic agents	Side	Clinical presentation	Location	Borden classification	Cognard classification	Arterial access	Arterial diameter (mm)	Venous access	Venous diameter (mm)	Material	Number of procedures performed per patient	Outcome	Complication
1	77	F	25.0	Yes	Right	Trias	CS	Ⅰ	II a	Proximal RA	2.2	Forearm CV	2.8	Coil	1	Subtotal occlusion	None
2	74	F	24.2	No	Left	Trias, neurological deficits (III, VI)	CS	Ⅱ	II a + b	Proximal RA	2.4	MCV	2.9	Coil	1	Total occlusion	None
3	79	F	35.2	No	Left	Trias, neurological deficits (III, IV, VI)	CS	II	II a + b	Distal RA	2.2	MCV	2.8	Coil	1	Total occlusion	None
4	81	F	22.2	Yes	Right	Chemosis	CS	Ⅰ	Ⅰ	Distal RA	2.1	Forearm CV	2.8	Coil	1	Total occlusion	None
5	59	M	22.5	No	Left	Trias, neurological deficit (III)	CS	Ⅱ	II b	Proximal RA	2.4	MCV	3.2	Coil	1	Subtotal occlusion	Subcutaneous hematoma at arterial puncture site
6	66	M	23.8	No	Left	Tinnitus	TSS	Ⅰ	II a	Distal RA	2.5	MCV	3.0	Coil	1	Total occlusion	None
7	73	M	22.6	Yes	Left	Hemorrhage and seizure	TSS	Ⅲ	Ⅲ	Distal RA	2.3	MCV	3.2	Coil	1	Total occlusion	None
8	79	F	23.4	Yes	Left	Trias, neurological deficit (III)	CS	Ⅰ	II a	Distal RA	2.1	MCV	3.0	Coil	1	Total occlusion	None
9	64	F	23.2	No	Right	Tinnitus	TSS	Ⅰ	II a	Proximal RA	2.2	Forearm CV	2.8	Coil	1	Total occlusion	None
10	48	M	22.4	No	Left	Trias, neurological deficits (III, VI)	CS	II	II a + b	Distal RA	2.4	MCV	2.9	Coil	1	Total occlusion	None
11	28	M	21.5	No	Left	Tinnitus	TSS	II	II a + b	Distal RA	2.3	MCV	3.2	Coil	1	Total occlusion	None
12	65	F	23.6	No	Right	Trias, neurological deficits (III, VI)	CS	II	II a + b	Distal RA	1.8	MCV	2.8	Coil	1	Total occlusion	None
13	75	F	23.4	Yes	Left	Tinnitus and hemorrhage	TSS	II	II a + b	Distal RA	2.0	Forearm CV	2.9	Coil	1	Total occlusion	None

CS, cavernous sinus; CV, cephalic vein; F, female; M, male; MCV, median cubital vein; RA, radial artery; Trias, chemosis, ocular bruit, and proptosis; TSS, transverse-sigmoid sinus.

Arterial puncture was performed at the distal radial artery in 9 patients. The right median cubital vein was punctured in 9 patients and the forearm cephalic vein in 4 patients. The mean diameter of punctured radial arteries and forearm veins are 2.2 and 2.9 mm, respectively. In all the patients, the transvenous guiding sheath was successfully inserted into the targeted vessel distal to the internal jugular vein. The angiographic result was total occlusion in 11 patients and subtotal occlusion in 2 patients. The procedure was completed in a single session in all the patients.

No instances of cerebral hemorrhage, ischemia, arterial puncture, or venous occlusion were observed during the follow-up period. Among the five patients receiving antithrombotic therapy, no complications at the puncture site were observed. One patient experienced a minor complication: a subcutaneous hematoma at the arterial puncture site, which required extended compression for resolution.

No patients reported procedure-related discomfort. Among the nine patients who had previously undergone transfemoral cerebral angiography, all reported experiencing less postprocedural discomfort with the current procedure.

Case presentation

Case 3

A 79-year-old woman with a high body mass index (35.2 kg/m²) presented with left proptosis, chemosis, subconjunctival hemorrhage, and disturbance of eye movement. Diagnostic transfemoral cerebral angiography revealed a dAVF in the left cavernous sinus. At the time, arterial puncture was challenging due to obesity. Subsequently, transvenous coil embolization using fTAVA was performed. Ultrasound-guided access to the right distal radial artery was achieved, and a 4-French Simmons guiding sheath was inserted. A forearm angiogram with a delayed venous phase helped cannulate the 4-French guiding sheath into the medial cubital vein. Subsequently, the transarterial and transvenous guiding sheaths were easily navigated to the left internal carotid artery and jugular bulb, respectively. The microcatheter was advanced into the left cavernous sinus through a coaxial catheter, and the left cavernous sinus and draining veins were embolized using coils. Finally, the fistula completely disappeared. The puncture site of the distal radial artery was occluded using a compression device (PreludeSYNC DISTAL; Merit Medical, South Jordan, UT, USA) and that of the medial cubital vein with bandage compression (Figure 1).

Figure 1.

Case 3. Forearm-only transarterial and venous approach for a patient with dural arteriovenous fistula in the left cavernous sinus. Right distal radial artery approach utilizing a 4-French guiding sheath (white arrowhead) and median cubital venous approach (black arrowhead) using the same set-up (a). A forearm angiogram with delayed venous phase helps to cannulate the 4-French guiding sheath into the median cubital vein (b). Upper arm venographic mapping helps navigate the transvenous catheter system into the subclavian vein (c). A second venogram is performed with the arterial catheter positioned at the left common carotid arteries (d, black arrowhead), clearly visualizing the left jugular and innominate veins and facilitating transvenous guidewire and JBII coaxial catheter guidance (d, white arrowhead). Left internal carotid angiogram showing a left cavernous sinus dural arteriovenous fistula with draining veins of the superior ophthalmic and uncal veins (e). Coil embolization at the cavernous sinus is performed using a microcatheter (Excelsior SL-10; Stryker Neurovascular, Fremont, CA, USA) passed through a 4-French TEMPO coaxial catheter (Cordis, Johnson & Johnson, Fremont, CA, USA) located at the inferior petrosal sinus (f). Postoperative angiogram revealing a complete disappearance of the dural arteriovenous fistula (g). Unobstructed compression is achieved using a distal radial compression device (PreludeSYNC DISTAL; Merit Medical, South Jordan, UT, USA) for closing the arterial puncture site, while bandage compression is used for closing in the right medial cubital vein approach (h).

Case 9

A 64-year-old woman presented with a right pulsatile tinnitus. Diagnostic transfemoral cerebral angiography revealed a dAVF in the right transverse-sigmoid sinus. Ultrasound-guided access to the right proximal radial artery was achieved, and a 4-French sheath was inserted. Subsequently, a 4-French Simmons catheter was advanced into the right common carotid artery after performing forearm angiogram. The cephalic vein of the right forearm was safely punctured under venographic mapping and ultrasound guidance. A transvenous 4-French guiding sheath was cannulated and navigated to the jugular bulb without difficulty. The microcatheter was advanced to the right transverse-sigmoid sinus through the coaxial catheter, and the shunt segment of the sinus was completely embolized with coils. The puncture site of the proximal radial artery was occluded using a compression device (TR Band; Terumo, Tokyo, JPN) and that of the forearm cephalic vein was occluded using bandage compression (Figure 2).

Figure 2.

Case 9. Forearm-only transarterial and venous approach for a patient with dural arteriovenous fistula in the right transverse-sigmoid sinus. A 4-French arterial sheath is located in the proximal radial artery, and a 4-French guiding sheath is inserted into forearm cephalic vein (a). Upper arm venographic mapping helps navigate the catheter system into subclavian vein (b). A second venogram is performed with the arterial catheter and the venous catheter simultaneously positioned at the right external carotid arteries (black arrowhead) and at the subclavian vein (white arrowhead), respectively. The jugular, subclavian veins, and superior vena cava are clearly visualized, and transvenous JBII coaxial catheter guidance is facilitated (c and d). The Excelsior SL-10 and guidewire are inserted into the transverse-sigmoid sinus (e, black arrowhead) using the guiding sheath located at the sigmoid sinus (e, white arrowhead), and coil embolization is successfully performed (f). Unobstructed compression is achieved using a proximal radial compression device (TR Band; Terumo, Tokyo, JPN) for closing the arterial puncture site, while bandage compression is used to close in the forearm cephalic vein approach.

Discussion

The present study confirmed the safety and effectiveness of fTAVA as a novel approach for endovascular intervention of dAVFs requiring both arterial and venous approaches.^3–6 This approach could be easily accomplished using standard neuroendovascular devices. Access to the distal left or right internal jugular vein could be successfully performed through the right forearm venous route using the JBII coaxial catheter, as commonly used in neurointervention.

Studies on the radial artery approach have shown a clear benefit over the transfemoral approach, with evidence of decreased morbidity and mortality, faster patient recovery time, higher post-procedure patient satisfaction, and even lower costs in the coronary interventional field.^7,8 Several recent investigations have confirmed the safety of transradial artery access in neurointervention.^9–11 Greater operator experience and the use of ultrasound-guided puncture are linked to lower risks of cannulation failure, fewer transfemoral crossovers, and decreased average contrast medium and fluoroscopic time.^9,12

Transvenous access for treating dAVFs is typically performed through puncture of the direct jugular or femoral veins.^3–6,13 Although rare, serious complications such as inadvertent arterial puncture (0.1%–1.6%), arterial canulation (0.3%), pneumothorax (0.1%–0.4%), hemothorax, arteriovenous fistula, arrhythmia, rupture of right atrium, air emboli, brachial plexus injury, vocal cord paralysis, catheter knotting, and severe respiratory obstruction have been reported in cannulations of the jugular or femoral vein (based solely on case reports).^14,15

These risks could increase in patients with severe obesity or in those taking oral antithrombotic drugs. ¹⁶ We recognize that the potential advantages of fTAVA over transfemoral methods remain theoretical as the complication rates of femoral and internal jugular vein cannulations under ultrasound guidance are quite low. ¹⁴ Although some of the aforementioned complications may also arise with fTAVA, we hypothesized that they could be managed more easily than with conventional methods. In previously reported large case series on forearm venous cannulation, serious complications including nerve injury and thrombosis have not been observed. ²

An alternative approach is required when the transfemoral approach is not suitable because of bilateral lower limb thrombosis, local groin infection, a dialysis catheter with exhausted venous access, central vein stenosis, or an inferior vena cava filter. In such cases, the internal jugular, superior ophthalmic, and superficial veins of the upper limbs (basilic, cephalic, and median cubital veins, respectively) are alternatives to the common femoral vein. Among them, the median cubital vein, located over the antecubital fossa, is a readily accessible and palpable vessel that drains into the basilic vein.^1–3 Although the basilic vein, with its straighter course and larger diameter, is the most commonly chosen puncture site,^1–3 it runs down the inside of the upper arm; thus, it is technically difficult to cannulate the catheters and to compress for hemostasis. Moreover, because the cephalic vein in the upper arm drains into the axillary vein at a steep angle, it is unsuitable for cannulation. However, the cephalic vein in the forearm can be easily cannulated, and catheters can be navigated through the median cubital vein. This suggests that the median cubital and forearm cephalic veins are preferable for the cannulation of the superficial forearm veins. The deep venous system of the forearm, which includes the radial, ulnar, and interosseous veins, is not an ideal access site due to the difficulty of puncture. ³

Abecassis et al. reviewed the clinical results of 147 attempted upper extremity transvenous approaches at 13 centers, including 32 dAVFs treated with transvenous embolization and numerous diagnoses. Although they also emphasized the accessibility of this method, 2 minor complications (1.4%) and 5 conversions to a transfemoral vein approach (3.4%) were reported. The cases requiring conversion included one instance of inadequate visualization of a suitable vein, one failed cephalic vein puncture, one accidental brachial artery puncture during an attempt to access the brachial vein, one case of sheath kinking or occlusion, and one failed distal catheterization. ² To avoid these complications, we recommend using ultrasound examination to confirm the route of the targeted vein and ensure safe cannulation. In a case series reported by Ramos et al., 28 patients underwent intracranial venography and venous stenting using the superficial upper extremity vein approach under ultrasound guidance without any complications. ¹ Thus, we also recommend the use of venography through the radial arterial sheath to confirm the diameter, course, and flow of the targeted vein.

As there are few reports on upper extremity transvenous approaches for dAVFs, a novel method of catheterization has not been standardized.^4–6 The approach to the left internal jugular vein is thought to be anatomically challenging because of the necessity to pass through the innominate vein between the bilateral superior vena cavae. Most previous reports have described the utility of the Simmons catheter as a coaxial catheter for cannulation of the targeted jugular vein.^1,4–6 Given that the shape of the Simmons catheter might make it difficult to use in venous vessels, we routinely used a 4-French JBII coaxial catheter, and this enabled easy and successful navigation to the targeted jugular vein in all cases.

The advantage of fTAVA over the transfemoral or direct jugular approach is that vessel puncture and hemostasis are easy and safe because the forearm veins have a shallow subcutaneous layer. Regarding catheter navigation, fTAVA requires downward movement at the subclavian vein or superior vena cava and return upward to the jugular vein. However, the arteriovenous shunt blood flow thickens the targeted vein, clarifies venography, and allows the guidewire to be guided sufficiently distally to follow the catheter.

With regard to patient satisfaction, fTAVA was associated with significantly less discomfort than the transfemoral approach. There are currently gold standard approaches for intracranial venous lesions, and only a few patients have been treated with fTAVA. However, fTAVA could be an alternative approach for patients at high risk of jugular or femoral vein puncture, with severe obesity, or taking oral antithrombotic agents.

Limitations

There are several limitations to this study. First, it is a retrospective study at a single facility, and the number of cases is small. Although this study reports only one minor complication, the small sample size may lead to an underestimation of the true complication rate. Moreover, this study does not include a control group including the patients underwent transfemoral approach or transarterial fistula occlusion with liquid embolic material, and does not address the learning curve associated with this fTAVA technique. To address these issues, a larger-scale study is warranted to more precisely estimate the efficacy and safety of the fTAVA. Second, the effectiveness of this method in patients with the venous stenosis, tortuosity, or a persistent left superior vena cava, which lacks a left innominate vein and occurs with a frequency of 0.3% to 0.5%, ¹⁷ remains unclear. In such cases, the right forearm transvenous approach may present anatomical challenges due to the low position of the bilateral pathway. Preprocedural evaluation of the left innominate vein using magnetic resonance angiography is essential, and the approach site may be changed to another location.

Conclusion

In this study, fTAVA for dAVF was successfully performed with a high success rate and no clinical serious complications. These results suggest that this method could be a safe and effective treatment option for patients with lesions distal to the right or left jugular vein.

Ethical statement

Ethical approval

This study was approved by the Institutional Review Board of St Marianna University School of Medicine, approval number 5301. Written informed consent was obtained from all participants included in this study.

ORCID iD

Hidemichi Ito https://orcid.org/0000-0002-1574-1894

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.*