Abstract
Background:
Neuroendovascular treatments, particularly transvenous embolization, require precise visualization of the approach route during microcatheter navigation. However, conventional roadmaps often face limitations, such as contrast medium dilution and slow venous flow. The multimodal image fusion guidance system integrates pre-acquired imaging datasets, such as computed tomography venography, with real-time fluoroscopy to enhance the visualization of the vascular anatomy and procedural accuracy. This report highlights the efficacy of this system in navigating microcatheters to complex arteriovenous fistulas (AVFs) using a transvenous approach.
Case Description:
Case 1 involved a 75-year-old woman with a left cavernous sinus dural AVF, which was treated using a transvenous approach through the facial vein. The multimodal image fusion guidance system enabled precise catheter navigation, resulting in complete obliteration of the shunt with coils. Case 2 involved a 45-year-old man with a spinal epidural AVF. This system facilitated accurate catheterization of the epidural venous plexus through the hemiazygos vein, enabling successful embolization with coils.
Conclusion:
The multimodal image fusion guidance system enhances visualization and catheter navigation, addressing the challenges inherent in transvenous procedures. These cases demonstrate its efficacy in improving procedural safety and precision in complex neurovascular conditions. Although limitations include dependence on pre-acquired images and potential misalignment due to vessel shifts, this system represents a significant advancement in neuroendovascular navigation. Further studies are warranted to evaluate its broader applicability and impact on clinical outcomes.
Keywords: Catheter navigation, Dural arteriovenous fistula, Multimodality image fusion guidance, Neuroendovascular treatment, Transvenous embolization
INTRODUCTION
Neuroendovascular treatment relies heavily on precise navigation, with accurate segmentation playing a crucial role in guiding the catheter toward the target lesion. Conventional roadmap guidance is widely used in trans-arterial approaches. However, in the case of the transvenous procedures, conventional roadmaps generated through contrast injection often face limitations such as contrast medium dilution and slow venous flow. These issues compromise roadmap quality and make catheter navigation more challenging during transvenous procedures.
Recent advancements in multimodal image fusion guidance have introduced innovative approaches for roadmap acquisition in neuroendovascular procedures.[3] This technology enables the integration of various imaging modalities, including pre-acquired datasets such as computed tomography venography (CTV) or magnetic resonance venography (MRV), with real-time two-dimensional (2D) fluoroscopic images. By creating a precise three-dimensional (3D) roadmap dynamically overlaid onto live imaging, this system enhances visualization and provides real-time guidance during catheter navigation. Previous reports have demonstrated the usefulness of image fusion techniques, such as fusing CT angiography or MR angiography with digital subtraction angiography (DSA), in endovascular aortic repair or intracranial arterial interventions, improving navigation accuracy and reducing contrast use.[8,9] The flexibility of multimodality fusion allows its application across diverse imaging datasets, thereby improving procedural accuracy and efficiency. Notably, this technology has the potential to overcome many limitations associated with transvenous approaches, offering improved visualization of the venous anatomy and facilitating safer and more precise catheter navigation.
Although some studies have described the application of image fusion in venous procedures, such reports remain limited in number, and particularly, few have demonstrated the usefulness of this technique for navigation during the treatment of arteriovenous malformation or dural arteriovenous fistulas (AVF). Because venous structures often exhibit greater anatomic variability and slower flow dynamics than arteries, visualization during venous navigation poses unique challenges that may be overcome by this technology.
In this report, we present two cases in which transvenous embolization (TVE) for cavernous sinus dural AVF and spinal epidural AVF was successfully performed using multimodal image fusion guidance, demonstrating its effectiveness as an alternative to traditional angiographic techniques in neuroendovascular treatment.
Method
The multimodal image fusion guidance used in this study was implemented using the Philips Vessel Navigator System equipped on an Azurion machine (Philips, Amsterdam, Netherlands). Preoperative contrast-enhanced CTV images were acquired and utilized for intraoperative guidance. These high-resolution venographic datasets were co-registered and overlaid onto real-time 2D fluoroscopic images during the procedure, enabling precise visualization of the vascular anatomy and accurate catheter navigation to the target lesion.
CASE PRESENTATION
Case 1: Cavernous sinus dural AVF
A 75-year-old woman presented to a referring hospital with complaints of lacrimation in her left eye. Examination revealed conjunctival hyperemia in the left eye. MR imaging (MRI) led to the diagnosis of a left cavernous sinus dural AVF [Figure 1].
Figure 1:

Preoperative magnetic resonance imaging (MRI) of Case 1. (A) Time-of-flight MR angiography (MRA) image showing an enlarged left superior ophthalmic vein (SOV) (arrow), (B) Time-of-flight MRA image at the level of the cavernous sinus showing no apparent abnormal high signal within the cavernous sinus (dotted arrow), (C) fluid-attenuated inversion recovery image also demonstrating the enlarged left SOV (arrow).
Cerebral angiography revealed a dural fistula at the anterior part of the cavernous sinus, fed by the left internal carotid artery meningohypophyseal trunk and draining into the left superior ophthalmic vein (SOV) [Figure 2]. In addition, minor shunt flow was identified from the left accessory and middle meningeal arteries. Drainage into the ipsilateral or contralateral inferior petrosal sinuses (IPSs) was not observed. Furthermore, conventional 2D angiography failed to adequately evaluate the venous drainage pathway distal to the SOV.
Figure 2:

Preoperative digital subtraction angiography of Case 1. (A) Frontal view, arterial phase. (B) Lateral view and arterial phase. Early drainage into the left superior ophthalmic vein (SOV) is observed (arrow). (C) Frontal view and venous phase. Venous drainage from the SOV is directed into the contralateral facial vein (arrowhead). The ipsilateral facial vein is not visible. (D) Lateral views, venous phase. The arrowhead indicates the contralateral facial vein.
The CTV demonstrated a venous drainage pathway extending from the left SOV to the left common facial vein.
Endovascular treatment was planned using a transvenous approach through the ipsilateral facial vein to achieve selective coil placement at the shunt point. The multimodal image fusion guidance system was employed by fusing the preoperative CTV data with real-time 2D fluoroscopy to assist navigation through the venous system [Figure 3A]. The devices used included a guiding catheter (FUBUKI 6Fr 90 cm AN, Asahi Intecc, Aichi, Japan), distal access catheter (Guidepost 3.2Fr-3.4Fr, Tokai Medical Products, Aichi, Japan), microcatheter (Marames 1.5-2.7Fr 165 cm, Medtronic, Minneapolis, MN, USA), and microguidewire (CHIKAI X 010 200 cm, Asahi Intecc, Aichi, Japan).
Figure 3:

Transvenous embolization procedure in Case 1. (A) Vessel navigator was used to overlay the left internal jugular vein and facial vein onto the fluoroscopic image. The microcatheter was navigated through the facial vein to the shunt point. The dotted arrow indicates the microcatheter tip, the solid arrow indicates the distal access catheter tip, and the arrowhead marks the microcatheter’s trajectory. (B and C) The shunt point is embolized with coils, resulting in complete occlusion of the shunt flow.
The guiding catheter was initially advanced to the confluence of the left common facial and internal jugular vein. Subsequently, the microcatheter and micro-guidewire were used to select the anterior facial vein, and the microcatheter was navigated through the SOV and into the shunt pouch. At the shunt pouch, 13 i-ED coils (COMPLEX SilkySoft, Kaneka, Osaka, Japan) were deployed with gradual downsizing to optimize the packing density and achieve complete shunt flow obliteration [Figures 3B and C]. Post-treatment angiography confirmed the disappearance of the shunt blood flow while preserving normal venous outflow from the SOV to the inferior ophthalmic vein. Microcatheter navigation using the multimodal image fusion guidance is shown in Figure 4. Preoperative CT venography was fused with real-time fluoroscopy, allowing advancement of the microcatheter from the anterior facial vein into the SOV and further to the shunt pouch. The arrow indicates the tip of the microcatheter, and the arrowhead indicates the tip of the micro guide wire.
Figure 4:

Representative fluoroscopic images demonstrating microcatheter navigation using the multimodal image fusion guidance in Case 1. Preoperative CT venography was fused with real-time fluoroscopy to visualize the venous anatomy. (A-C) The image demonstrates navigation of the microcatheter from the anterior facial vein into the SOV and subsequently to the shunt pouch.The arrow indicates the tip of the microcatheter, and the arrowhead indicates the tip of the micro guide wire.(D) Angiographic image obtained after advancement to just proximal to the shunt pouch.
The patient’s symptoms, including lacrimation and conjunctival hyperemia, improved significantly after the procedure. The patient was discharged in good condition and remained symptom-free without recurrence for 1 year after the procedure.
Case 2: Spinal epidural AVF
A 45-year-old man presented to a referring hospital with subacute difficulty in walking. MRI suggested the presence of a spinal epidural tumor. The referring institution attempted posterior fixation from Th2 to Th6 and laminectomy from Th3 to Th5, along with biopsy. However, the procedure was complicated by significant bleeding, resulting in the completion of posterior fixation alone. Given the suspicion of spinal epidural AVF, the patient was referred to our institution for further evaluation and management.
On admission to our hospital, the patient exhibited bilateral lower limb paralysis; sensory deficits to pain, temperature, and touch below the Th2 level; and bladder and bowel dysfunction. MRI demonstrated extensive T2-weighted hyperintensity in the thoracic spinal cord, indicative of aggressive venous congestion [Figure 5].
Figure 5:

Preoperative magnetic resonance imaging of Case 2. (A) Sagittal T2-weighted image of the thoracic spinal cord showing high intramedullary signal intensity, suggesting venous congestion (arrow). Dilated spinal veins are also visualized as flow voids (dotted arrow). (B) Axial image at the Th7 level also showing high signal intensity in the spinal cord, suggesting venous congestion (arrow).
Spinal angiography revealed multiple diffuse spinal epidural AVFs from Th4 to Th7. These AVFs drain into the epidural venous plexus, with the flow ultimately directed through the hemiazygos vein and inferior vena cava (IVC) [Figure 6]. Although no clear intradural drainage was observed on angiography, marked T2 hyperintensity on MRI prompted further investigation using a microcatheter. To prevent further venous congestion and potential progression of spinal cord dysfunction, a combined approach involving transarterial embolization followed by transvenous coil embolization of the epidural venous plexus was planned.
Figure 6:

Preoperative digital subtraction angiography of Case 2. (A) Left Th7 radicular artery angiography, arterial phase. The epidural venous plexus (thin arrow) and hemiazygos vein (dotted arrow) are visualized. (B) Left Th7 radicular artery angiography in the arterial phase. The superior vena cava was visualized (thick arrow).
A transvenous approach was used through the femoral vein and navigated through the IVC, superior vena cava, and hemiazygos veins. The system employed a variety of devices, including the guiding catheter (BENCHMARK 6Fr 95 cm, Penumbra, Alameda, CA, USA), distal access catheter (Guidepost 3.2Fr-3.4Fr 120 cm, Tokai Medical Products, Aichi, Japan), microcatheter (Headway Duo 156 cm, MicroVention, Aliso Viejo, CA, USA; DeFrictor nano 1.3/2.2Fr 165 cm, Asahi Intecc, Aichi, Japan), and microguidewire (Traxcess 012/014, MicroVention, Aliso Viejo, CA, USA; CHIKAI X010, Asahi Intecc, Aichi, Japan). Diagnostic catheters were placed in the left Th4/5 and right Th5/6 segmental arteries through the bilateral femoral arteries.
Using the multimodal image fusion guidance system, pre-acquired CTV data were fused with real-time 2D fluoroscopy, creating a roadmap that facilitated navigation of the guiding catheter to the hemiazygos vein [Figure 7A]. The spinal vein at the L3 level was selected and the Guidepost/Headway Duo microcatheter was advanced into the epidural venous plexus. Up onto the Th5 level [Figure 7B], a total of nine coils were deployed, successfully reducing flow into the hemiazygos vein [Figure 7C]. Final angiography confirmed a significant reduction in shunt flow.
Figure 7:

Transvenous embolization procedure in Case 2. (A) Vessel navigator was utilized to overlay the inferior vena cava, superior vena cava, and hemiazygos vein onto the fluoroscopic image, facilitating guidance of the catheter to the hemiazygos vein. (B) The microcatheter is navigated from the hemiazygos vein to the epidural venous plexus. The dotted arrow indicates the microcatheter tip, solid arrow indicates the distal access catheter tip, and arrowhead indicates the trajectory of the microcatheter. (C) Fluoroscopic image demonstrating coil embolization of the epidural venous plexus.
The patient did not experience any postoperative neurological deficits. Postoperative MRI showed a slight improvement in the enlargement of the spinal veins and a reduction in spinal cord swelling. The patient’s lower limb paralysis and sensory deficits remained unchanged compared to preoperative levels. The patient was then transferred to a rehabilitation facility for continued gait disturbance rehabilitation.
DISCUSSION
TVE has become an effective and often preferred treatment for dural and spinal epidural AVFs.[4,6] For intracranial dural AVFs, particularly those involving challenging locations such as the cavernous sinus and hypoglossal canal, TVE is considered the first-line approach, achieving angiographic cure rates of 80–90%.[1,6] However, veins exhibit greater anatomical variability than arteries,[5,7,10] making a thorough understanding of the venous anatomy crucial for successful catheter navigation. Consequently, the success of TVE relies heavily on precise venous segmentation, which is essential for safe and accurate guidance of microcatheters to the target lesion. Conventional DSA is frequently used to guide navigation; however, it has several limitations in depicting venous anatomy with clarity. Issues, such as contrast dilution and slow flow rates through venous structures, impede the accurate visualization of complex or tortuous venous systems, adding complexity to the navigation required in TVE procedures and limiting the efficacy of treatment. [3] The inherent challenges of DSA-based imaging underscore the need for enhanced visualization technologies to improve safety and precision of TVE.
The multimodal image-fusion guidance system presents a promising solution to these challenges. This system works by integrating high-resolution preoperative imaging data, such as CT or MRI, with real-time 2D fluoroscopic images, creating an overlay that offers a precise 3D roadmap of the vascular anatomy within the fluoroscopic view. Once the 3D roadmap was created and co-registered with 2D fluoroscopy, it was preserved even after C-arm rotation and/or table movement, avoiding the repetitive time and contrast agent consumption for roadmap acquisition in each position. In abdominal endovascular procedures, studies have reported a reduction in contrast agent usage of as much as 70%[9] and a decrease in procedural time of 18%[8] when using image fusion guidance. Such benefits suggest that image fusion technology not only enhances procedural efficiency but also significantly reduces patient exposure to contrast agents, which is a crucial consideration in complex neurovascular cases that often require prolonged imaging and extensive contrast administration.
In the first case, we chose a transvenous approach through the facial vein. While the IPS is commonly used as the preferred route for TVE in cases of cavernous sinus dural AVF, transvenous access through the facial vein has been reported in cases in which the IPS is thrombosed.[2] However, the catheter navigation may be more complex in such cases. The multimodal image fusion guidance system facilitated precise navigation, ensuring a safe and effective procedure even in cases where thrombosis of the IPS made guidance, such as approach through the facial vein, challenging. In the second case, involving a spinal epidural AVF, TVE was commonly reported as a treatment option.[4] Given that catheter navigation to the epidural venous plexus can be challenging, the system supports precise navigation through the hemiazygos vein, ultimately facilitating embolization. The ability to overlay 3D venographic imaging over real-time fluoroscopy facilitates accurate transvenous access to the target site, enabling successful embolization of the shunt.
In addition to its use in TVE for cavernous dural AVF and spinal dural AVF, as demonstrated in the present cases, the multimodality image fusion guidance system has also been shown to improve sinus stenting for transverse sinus stenosis, as reported in the literature.[3] Together with these previous reports, the current case further emphasizes the effectiveness of the system in complementing conventional angiographic techniques.
In addition, it minimizes the potential risks associated with traditional guidance techniques. By providing a more reliable roadmap, this technology may reduce the likelihood of navigational errors and procedural complications, thereby improving the overall success rate of neuroendovascular procedures. The ability to reduce contrast agent use and fluoroscopy time, as seen in other vascular applications, could translate into reduced patient exposure to ionizing radiation and nephrotoxic agents, which is particularly important for patients with renal impairment or those undergoing multiple interventions.[8,9] For patients with complex neurovascular conditions, reduced contrast load and fluoroscopic exposure may enable safer and more effective long-term management of these challenging cases.
Despite the demonstrated benefits of the proposed multimodal image fusion guidance system, some limitations must be acknowledged. First, because the system relies on the fusion of pre-acquired imaging data, intraoperative modifications or updates to the 3D roadmap are challenging, potentially limiting its adaptability to dynamic procedural environments. In particular, during cervical vein segmentation, variations in the degree of cervical flexion and extension between the treatment and imaging phases may result in ineffective segmentation. Second, catheter manipulation during the procedure can induce vessel shifts, leading to a misalignment between the real-time anatomy and pre-acquired 3D roadmap, which may affect the accuracy of navigation and require additional adjustments by the operator. These factors highlight the need for further advancements in real-time imaging integration and adaptive guidance technology to optimize the efficacy of the system in endovascular interventions.
This report emphasizes the efficacy of a multimodal image fusion guidance system in TVE for treating complex vascular lesions, including dural and spinal epidural AVF. The system significantly enhances the visualization and precision of catheter navigation, contributing to successful treatment outcomes in the presented cases. Further research is essential to fully assess its impact on procedural parameters and overall patient safety during neuroendovascular interventions.
CONCLUSION
The multimodal image fusion guidance system provides a reliable and practical method for enhancing visualization and accuracy during TVE. By integrating pre-acquired venographic datasets with real-time fluoroscopy, this technique allows precise microcatheter navigation even in complex or tortuous venous pathways where conventional roadmaps are limited. In the present two cases of cavernous sinus dural AVF and spinal epidural AVF, the system facilitated safe and effective catheterization, leading to successful embolization. Although its current use depends on preoperative imaging and may be affected by vessel shifts, multimodal image fusion guidance represents a promising advancement in neuroendovascular navigation, warranting further evaluation in larger clinical series.
Footnotes
How to cite this article: Fujii G, Koizumi S, Fujitani S, Umekawa M, Ono H, Saito N. Multimodality image fusion guidance in the microcatheter navigation to cerebrospinal arteriovenous fistula: Report of the two cases. Surg Neurol Int. 2026;17:133. doi: 10.25259/SNI_995_2025
Contributor Information
Genta Fujii, Email: gfujii917@gmail.com.
Satoshi Koizumi, Email: sakoizumi-tky@umin.net.
Shigeta Fujitani, Email: sfujitani@gmail.com.
Motoyuki Umekawa, Email: moto.umekawa@gmail.com.
Hideaki Ono, Email: onoh@m.u-tokyo.ac.jp.
Nobuhito Saito, Email: nsaito-nsu@m.u-tokyo.ac.jp.
Ethical approval:
The Institutional Review Board approval is not required.
Declaration of patient consent:
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patients have given their consent for their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship:
Nil.
Conflicts of interest:
There are no conflicts of interest.
Use of artificial intelligence (AI)-assisted technology for manuscript preparation:
The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.
Disclaimer
The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Journal or its management. The information contained in this article should not be considered to be medical advice; patients should consult their own physicians for advice as to their specific medical needs.
REFERENCES
- 1.Bhatia KD, Lee H, Kortman H, Klostranec J, Guest W, Wälchli T, et al. Endovascular management of intracranial dural AVFs: Transvenous approach. AJNR Am J Neuroradiol. 2022;43:510–6. doi: 10.3174/ajnr.A7300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Biondi A, Milea D, Cognard C, Ricciardi GK, Bonneville F, Van Effenterre R. Cavernous sinus dural fistulae treated by transvenous approach through the facial vein: Report of seven cases and review of the literature. AJNR Am J Neuroradiol. 2003;24:1240–6. [PMC free article] [PubMed] [Google Scholar]
- 3.Blanc R, Fahed R, Roux P, Smajda S, Ciccio G, Desilles JP, et al. Augmented 3D venous navigation for neuroendovascular procedures. J Neurointerv Surg. 2018;10:649–52. doi: 10.1136/neurintsurg-2017-013365. [DOI] [PubMed] [Google Scholar]
- 4.Brinjikji W, Yin R, Nasr DM, Lanzino G. Spinal epidural arteriovenous fistulas. J Neurointerv Surg. 2016;8:1305–10. doi: 10.1136/neurintsurg-2015-012181. [DOI] [PubMed] [Google Scholar]
- 5.Dalip D, Iwanaga J, Loukas M, Oskouian RJ, Tubbs RS. Review of the variations of the superficial veins of the neck. Cureus. 2018;10:e2826. doi: 10.7759/cureus.2826. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Klisch J, Huppertz HJ, Spetzger U, Hetzel A, Seeger W, Schumacher M. Transvenous treatment of carotid cavernous and dural arteriovenous fistulae: Results for 31 patients and review of the literature. Neurosurgery. 2003;53:836–56. doi: 10.1227/01.neu.0000083551.26295.ab. discussion 856-7. [DOI] [PubMed] [Google Scholar]
- 7.Priya A, Philip SE, Jain A, Sikka A. Variations of azygos vein: A cadaveric study with clinical relevance. Anat Cell Biol. 2023;56:448–55. doi: 10.5115/acb.23.074. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Sailer AM, De Haan MW, Peppelenbosch AG, Jacobs MJ, Wildberger JE, Schurink GW. CTA with fluoroscopy image fusion guidance in endovascular complex aortic aneurysm repair. Eur J Vasc Endovasc Surg. 2014;47:349–56. doi: 10.1016/j.ejvs.2013.12.022. [DOI] [PubMed] [Google Scholar]
- 9.Tacher V, Lin M, Desgranges P, Deux JF, Grünhagen T, Becquemin JP, et al. Image guidance for endovascular repair of complex aortic aneurysms: Comparison of two-dimensional and three-dimensional angiography and image fusion. J Vasc Interv Radiol. 2013;24:1698–706. doi: 10.1016/j.jvir.2013.07.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zhou Z, Xu K, Yu J. Transfemoral trans-facial vein-superior ophthalmic vein to embolize cavernous sinus dural arteriovenous fistulas. Front Neurol. 2022;13:1078185. doi: 10.3389/fneur.2022.1078185. [DOI] [PMC free article] [PubMed] [Google Scholar]
