Current Insights and Management Strategies for Lower Cervical Arteriovenous Fistulas: A Comprehensive Review

Abstract

Lower cervical arteriovenous fistulas (AVFs) are rare and complex vascular malformations that pose significant clinical challenges due to their location and variable presentation. While upper cervical AVFs have been extensively studied, lower cervical AVFs remain underresearched. This study aims to review the clinical presentations, management strategies, and outcomes of patients with lower cervical AVFs to enhance understanding and improve treatment approaches. We conducted a retrospective analysis of patients with spinal vascular malformations treated at our institute between June 2006 and December 2023, identifying two cases of lower cervical AVFs. Additionally, a systematic literature review was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, including 44 patients with lower cervical AVFs, using databases such as Ovid MEDLINE, PubMed, and Cochrane. Data collected included patient demographics, clinical presentation, fistula type, arterial and venous involvement, treatment modality, and neurological outcomes. Among the 44 patients with lower cervical AVFs, including our two cases, 50% were female, and the mean age was 48.68 years (range: 4-76 years). Clinical presentations varied, with 27.3% experiencing hemorrhage, 18.2% presenting with myelopathy, and 18.2% remaining asymptomatic. Venous drainage patterns played a significant role in symptom severity, with complex perimedullary and retrograde venous drainage contributing to worse outcomes. Treatment included endovascular embolization (40.9%), surgical resection (25%), and combined approaches (18.2%), with good recovery achieved in 54.5% of cases. Lower cervical AVFs present diverse clinical challenges due to their variable presentations and complex vascular anatomy. Early diagnosis and tailored management, including endovascular embolization and surgical resection, are essential for optimizing patient outcomes. Further research is needed to better understand the natural history of asymptomatic AVFs and improve treatment protocols.

Keywords: epidural arteriovenous fistulas; lower cervical arteriovenous fistulas; perimedullary arteriovenous fistulas; radicular arteriovenous fistulas; spinal cord venous infarction; spinal dural arteriovenous fistula.

Fig. 1

Case illustration 1. Lower cervical epidural arteriovenous fistula. ( A ) Sagittal T2-weighted magnetic resonance imaging of the cervicothoracic spine showed abnormally high-signal intensity from the lower brainstem to the mid-thoracic cord, indicative of venous congestion, with intradural flow voids evident along both the anterior and posterior surfaces of the spinal cord. Anteroposterior views of ( B ) contrast-enhanced magnetic resonance angiography of brachiocephalic trunk and vertebrobasilar system, and ( C ) angiographic image, ( D ) three-dimensional (3D) reconstruction image, and ( E , F ) maximum intensity projection reformatted images of angiographic computerized tomography of the right costocervical trunk revealed a spinal epidural arteriovenous fistula along the C8 nerve (C7-T1 vertebral foramen). This fistula is supplied by an enlarged deep cervical artery with prominently engorged draining veins extending rostrally to the pial venous network around the medulla and caudally to the thoracic spinal cord. A venous pouch is indicated by arrowheads.

Fig. 2

Case illustration 1. Lower cervical epidural arteriovenous fistula. ( A ) Anteroposterior view of superselective angiographic image of the right deep cervical artery clearly demonstrated a spinal epidural arteriovenous fistula and venous pouch (arrowhead). ( B ) Image during embolization showed the glue penetrating the fistula and proximal vein. ( C ) Postembolization image from the right costocervical trunk injection confirmed complete obliteration of the fistula. Sagittal T2-weighted magnetic resonance imaging at ( D ) 4 months, ( E ) 2 years, and ( F ) 5 years after embolization revealed complete and lasting resolution of venous congestion. ( G ) Contrast-enhanced magnetic resonance angiography of brachiocephalic trunk and vertebrobasilar system obtained 5 years following embolization confirmed no recurrence of the fistula.

Fig. 3

Case illustration 2. Lower cervical dural arteriovenous fistula. ( A ) Sagittal T2-weighted magnetic resonance imaging of the cervical spine showed faint abnormally high-signal intensity of the cervical cord with subtle intradural flow voids evident along the posterior surface of the spinal cord. ( B ) Sagittal three-dimensional (3D) T2-SPACE image demonstrated more prominent of intradural flow voids (white arrowheads). ( C ) Coronal T2 turbo inversion recovery magnitude (TIRM) image showed intradural flow voids along the left-sided cervical cord. Anteroposterior (AP) views of ( D ) contrast-enhanced magnetic resonance angiography and ( E ) angiography of brachiocephalic trunk and vertebrobasilar system revealed a spinal dural arteriovenous fistula (arrow) along the C5 nerve supplied by the left ascending cervical artery with engorged draining veins extending rostrally to the pial venous network around the upper cervical cord. ( F ) Lateral view of the left ascending cervical artery injection disclosed dilated perimedullary vein (black arrowheads) along the left-side cervical cord. ( G ) AP view of the left ascending cervical artery injection in venous phase illustrated venous outlet pathways (asterisks) of the fistula via the right deep cervical and left C8 epidural veins.

Fig. 4

Case illustration 2. Lower cervical dural arteriovenous fistula. ( A ) Anteroposterior (AP) and ( B ) lateral views of the superselective left ascending cervical artery injection clearly showed a spinal dural arteriovenous fistula (arrow) along the C5 nerve with venous drainage into engorged draining veins extending rostrally and caudally along the posterior aspect of the cervical cord through a left-sided dilated perimedullary vein (arrowhead). The venous outlet pathways (asterisks) were noted at the right deep cervical and left C8 epidural veins. ( C ) During embolization, the glue cast penetrated into the fistula and draining vein. ( D ) AP and ( E ) lateral views of the left ascending cervical artery following embolization confirmed complete obliteration of the fistula. ( F ) Sagittal and ( G ) coronal views of Dyna computerized tomography image after glue embolization demonstrated the glue cast at the posterolateral aspect of the C4 cord.

Fig. 5

Case illustration 2. Magnetic resonance imaging obtained 1 day after embolization. Sagittal ( A ) T2-weight, ( B ) diffusion-weighted imaging (DWI), and ( C ) apparent diffusion coefficient (ADC) images of the cervical spine and axial ( D ) T2-wighted and ( E ) gradient-recalled echo (GRE) images of the C4 level revealed a left-sided focal hyperintense T2 lesion at the C3-C4 level of the cervical cord with restricted diffusion on DWI and ADC images, indicative of acute cervical cord ischemia.

Fig. 6

Case illustration 2. Magnetic resonance imaging obtained 5 months after embolization. Sagittal ( A ) T2-weighted, ( B ) diffusion-weighted imaging (DWI), and ( C ) apparent diffusion coefficient (ADC) images of the cervical spine and axial ( D ) T2-wighted and ( E ) gradient-recalled echo (GRE) images of the C4 level demonstrated a subtle residual hyperintense lesion on the left side of the cervical cord at the C4 level without restricted diffusion on DWI and ADC images.

Fig. 7

Flow diagram showing a summary of our search strategy using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for relevant studies on lower cervical arteriovenous fistulas.