2026 Volume 13 Pages 91-96
Although rare, thrombosed vertebral artery aneurysms can lead to severe symptoms and are challenging to treat due to their proximity to vital structures. The location of a thrombosed aneurysm on the anterior aspect of the brainstem poses a significant challenge to conventional microscopic approaches. We herein report a 78-year-old man with a thrombosed right vertebral artery aneurysm who developed progressive quadriparesis, dysphagia, and respiratory failure despite prior interventions, including flow diverter stent placement, parent artery occlusion, and microscopic thrombectomy. Given worsening medullary compression and poor clinical status, intra-aneurysmal thrombectomy was performed using an exo-endoscopic 2-step approach. The procedure involved reopening the previous suboccipital craniotomy, partial condylectomy, and C1 hemilaminectomy. Exoscopic thrombus de-bulking was followed by endoscopic evacuation of the residual thrombus compressing the ventral brainstem. Postoperatively, no complication was observed, and the patient demonstrated gradual neurological improvement, including recovery of spontaneous respiration and the ability to wean from mechanical ventilation within 3 weeks. Follow-up imaging confirmed resolution of medullary compression without thrombus recurrence. The exo-endoscopic 2-step approach is a viable option for surgical decompression of thrombosed vertebral artery aneurysms that cause brainstem compression. This enhances surgical access and visualization, particularly in the ventral brainstem, while potentially minimizing brainstem manipulation. Further investigation is warranted to better define the indications, efficacy, and safety of the management of complex thrombosed aneurysms.
Vertebral artery (VA) aneurysms represent <5% of intracranial aneurysms and are relatively uncommon vascular lesions. Anatomically, the VA is divided into 4 distinct segments: V1 (pre-foraminal), V2 (foraminal), V3 (C2-dura), and V4 (intradural).1) Aneurysms arising from the V4 segment pose unique management challenges because of their location and potential to compress the adjacent medulla oblongata.2) Compression of the medulla oblongata by these aneurysms manifests as a diverse range of atypical symptoms, including progressive quadriparesis, respiratory dysfunction, paraesthesia, dysarthria, visual disturbances, and cognitive changes.3) Conventional transcranial approaches are often hampered by deep and narrow surgical corridors that restrict instrument manoeuvrability and vascular control of the parent artery.4) The exo-endoscopic 2-step approach (EETA) has emerged as a less-invasive alternative for managing lesions at the skull base, including those at the craniovertebral junction. This technique enhances surgical field visualization with minimal invasiveness.5,6) Herein, we present a patient with a thrombosed right VA aneurysm who underwent intra-aneurysmal thrombectomy via the EETA aimed at relieving medullary compression and improving clinical status.
Radiological time course showing multiple endovascular treatments and the initial intra-aneurysmal thrombectomy. A, B DSA of the right vertebral artery before (A) and after (B) the initial stent placement. The aneurysm was thrombosed, with blood flow remaining only near the aneurysm neck. A stent was deployed to cover the neck. Red arrowheads outline the thrombosed aneurysm. C Follow-up 3D-DSA after stent placement. White arrowhead: site of blood flow recurrence at the aneurysm neck. White arrow, a perforator supplying the medulla. D DSA obtained after parent artery occlusion. White arrows, extent of coil embolization within the stent. E Gadolinium-enhanced T1-weighted MRI showing enlargement of the thrombosed aneurysm despite parent artery occlusion. F Follow-up DSA revealed a slight recurrence of blood flow at the aneurysm neck even after parent artery occlusion. G Skin incision design used for the initial intra-aneurysmal thrombectomy. H Gadolinium-enhanced T1-weighted MRI obtained immediately after the initial intra-aneurysmal thrombectomy. I, J Gadolinium-enhanced T1-weighted MRI and 3D computed tomographic angiography obtained 3 months after the initial intra-aneurysmal thrombectomy. Black arrowhead, thrombosed right vertebral artery; Blue arrowheads, the distal and proximal ends of the stent, the lumen of which was occluded by coils and thrombus; Light blue arrow, the distal end of the thrombosed segment of the right vertebral artery.
3D: 3-dimensional; DSA: digital subtraction angiography; MRI: magnetic resonance imaging
Schematic view and intraoperative pictures of thrombectomy using an exoscope. A Schematic view of thrombectomy using an exoscope. B Prior to manipulating the aneurysm wall, we confirmed sufficient working space to endoscopically observe and clip the vertebral artery distal to the aneurysm. C The inferolateral aspect of the aneurysm was exposed through the interval between the brainstem and the vertebral artery. The illustration on the bottom left indicates the field of view. White arrowhead, brainstem; Red arrowheads, aneurysm wall; Yellow arrowhead, vertebral artery. D The lateral part of the thrombus was removed using the cranial (blue dashed circle) and caudal (black dashed circle) corridors. The illustration on the bottom left indicates the field of view. White arrowhead, brainstem; Green arrowheads, intra-aneurysmal thrombus; Yellow arrowhead, vertebral artery.
Intraoperative endoscopic pictures. A Schematic view of thrombectomy using an endoscope. B Residual thrombus in the ventral aspect of the brainstem was removed using an endoscope. The illustration on the bottom left indicates the field of view. White arrowhead, brainstem; Blue arrowhead, caudal opening in the aneurysm wall; Yellow arrowhead, vertebral artery; Green arrowhead, thrombus. C, D Intra-aneurysmal view at the same place. The illustration on the bottom left indicates the field of view. C The majority of the thrombus was removed, with a small residual thrombus (green arrowhead) left around the orifice of the aneurysm. White arrow, the portion most deeply indenting the ventral medulla; Light blue arrowhead, the cranial opening of aneurysm (= blue dashed circle in Figure 2D). D ICG angiography. Green arrowhead, residual thrombus around the aneurysmal neck; Orange arrowhead, transmitted fluorescence from the left vertebral artery. ICG stains both the external (red arrowhead) and internal surface of the aneurysm wall.
ICG: Intraoperative indocyanine green
Postoperative CT and gadolinium-enhanced T1-weighted MRI. A Axial CT. Red arrowhead, stent placed in right VA; Green arrowhead, resected portion of the occipital condyle. B Posterior view of 3-dimensional computed tomography. Red arrowheads, stent placed in right VA; Green arrowheads, resected portion of the occipital condyle; Black arrowheads, articular surface of the occipital condyle; Black arrows, hypoglossal canal (the area posterior to the entrance is opened on the right). C Axial view of postoperative MRI day 7. D Axial view of postoperative MRI day 34.
CT: computed tomography; MRI: magnetic resonance imaging; VA: vertebral artery
This study was conducted in accordance with the guidelines of the Declaration of Helsinki. As this is a technical case report, approval from the Institutional Review Board of Nagoya University was not required. Written informed consent was obtained from the patient.
A 78-year-old male patient presented with respiratory failure requiring mechanical ventilation that developed 3 months before the current surgical intervention. Five years earlier, the patient underwent flow diverter stent (FDS) placement for an asymptomatic right VA large thrombosed aneurysm (Figure 1A and B). Although digital subtraction angiography (DSA) confirmed complete occlusion 1 year after FDS, the thrombosed aneurysm remained asymptomatic and enlarged slowly over time. Due to the onset of gait disturbances and mild dysphagia, careful observation with DSA revealed a faint vascular channel entering the thromboembolic aneurysm (Figure 1C). Despite the risk of medullary infarction, FDS was deemed ineffective, leading to parent artery occlusion (PAO) with coils (Figure 1D). Unfortunately, 3 months after the initial PAO, the aneurysm continued to grow progressively on magnetic resonance imaging (MRI) (Figure 1E), partly because of reopening of the occluded coil portion (Figure 1F), resulting in worsening medulla oblongata compression and the development of severe pneumonia. Given progressive respiratory and neurological deterioration, endovascular trapping of the aneurysm with a coil and n-butyl cyanoacrylate (NBCA) was followed by an initial microscopic intra-aneurysmal thrombectomy via a condylar fossa approach with a hockey stick-shaped skin incision (Figure 1G). However, due to insufficient visualization of the aneurysm during this procedure, only a small portion of the thrombus could be removed. Consequently, despite monitoring the patient for 3 months, this intervention yielded no clinical improvement. Subsequently, the thrombosed aneurysm continued to enlarge, based on the latest preoperative MRI (Figure 1H and I), the patient's spontaneous respiratory drive diminished further, and left-dominant quadriparesis worsened. Therefore, following FDS placement, the aneurysm enlargement was diagnosed as an extremely refractory aneurysm where residual mass effect persisted even after PAO and partial thrombus resection. The present surgical concept was to reduce the mass effect by safely removing the maximum possible thrombus. The clinical goals in this case were weaning from mechanical ventilation and improvement of neurological symptoms. Consequently, a second intra-aneurysmal thrombectomy was performed to decompress the brainstem and improve his respiratory and neurological status. 3-dimensional computed tomography (CT) performed prior to the second surgery demonstrated occlusion of the right VA both proximal and distal to the aneurysm, with no residual flow around the neck (Figure 1J). Given that the NBCA was localized to the immediate vicinity of the neck and the bulk of the intra-aneurysmal mass was pure thrombus, the intra-aneurysmal thrombus was deemed amenable to removal via dissection and suction using malleable instruments.
We elected to reutilize the condylar fossa approach from the initial operation to ensure proximal control of the VA while sparing the patient additional surgical trauma. Since conventional microscopic surgery had resulted in insufficient thrombus removal, our strategy for the second attempt involved a combined approach: utilizing the stereoscopic view of an exoscope and endoscopic visualization to expose deep regions and blind spots for more extensive evacuation. The procedure was conducted with continuous intraoperative monitoring of motor-evoked potentials from both vocal cords and extremities. The patient was placed in the prone position, and the previous craniotomy site was reopened. To facilitate access to the aneurysm by widening the space between the VA and the brainstem, the occipital condyle was partially removed. After dural opening, the inferolateral aspect of the aneurysm was exposed through the interval between the brainstem and VA using an exoscope (ORBEYE; Olympus, Tokyo, Japan) (Figure 2A). Prior to manipulating the aneurysm wall, we confirmed sufficient working space to endoscopically observe and clip the VA distal to the aneurysm (Figure 2B). An aneurysm wall opening was performed to allow thrombus de-bulking through the 2 corridors (Figure 2B and C). However, the residual thrombus located ventral to the brainstem was not visible on exoscopic examination. Due to the limited field of view provided by the exoscope, the procedure was continued using an endoscope. Rigid endoscopes (0°, 30°, or 70°; outer diameter, 4 mm; Karl Storz, Tuttlingen, Germany) were used to visualize and access the ventral aspect of the brainstem to remove the residual thrombi (Figure 3A-C). Curved surgical instruments were used to manipulate the interior of the aneurysm deeply embedded in the brainstem. We concluded the thrombectomy, leaving the thrombus adjacent to the neck in place, considering the possibility that it contained NBCA and was adherent to the stent. Although minimal blood flow was observed on the external and internal surfaces of the aneurysm wall, indocyanine green fluorescence endoscopy (0°; outer diameter, 4 mm; Karl Storz) confirmed the absence of bleeding inside the aneurysm (Figure 3D). The closure involved standard dural, fascial, and skin suturing, as well as bone-flap fixation using a titanium plate.
Postoperative non-contrast CT revealed no haemorrhage at the surgical site and preservation of most of the right occipital condyle (Figure 4A and B). Subsequently, the patient's quadriplegia gradually improved. The patient was weaned from mechanical ventilation 3 weeks postoperatively and no longer required supplemental oxygen. Follow-up postoperative MRI several days and 1 month after surgery showed no ischaemic or oedematous changes in the central nervous system, including the brainstem and cerebellum, nor was there any reaccumulation of thrombus within the aneurysm (Figure 4C and D). The patient was then referred for medical rehabilitation aimed at improving his physical condition and preventing further complications related to prolonged immobilization.
Thrombosed intradural VA aneurysms are rare and technically demanding to treat due to their proximity to the brainstem and cranial nerves. Despite their rarity, mass effects can lead to profound deficits.1) In the present case, progressive deterioration despite prior endovascular treatment justified decompressive surgery. Microsurgical approaches are reserved for symptomatic patients because of the depth and narrowing of the surgical corridors.4) The EETA is a viable option that was initially developed for anterior craniovertebral junction tumours.5,7) Although a microscope can be used instead of an exoscope during the initial phase, endoscopic visualization is superior for assessing the ventral aspect of the brainstem. It combines the broad visual field of the exoscope with the angled access of the endoscope, thereby overcoming individual limitations.5,6)
In our approach, a suboccipital craniotomy with partial condylectomy and C1 hemilaminectomy was performed, which is a well-validated route to the ventral brainstem. Exoscopy-guided thrombus de-bulking reduced lateral compression, whereas the endoscope enabled the safe removal of the residual ventral thrombus and ensured haemostasis. Compared to other techniques, such as the combined transcranial-endonasal approach described by Saito et al.,8) our single-position EETA allowed ergonomic advantages, uninterrupted neuromonitoring, and reduced surgical invasiveness.5) Yamamoto et al.9) reported a similar case of VA aneurysm-induced medullary compression treated using a far-lateral transcondylar approach. The patient showed clinical improvement, supporting surgical decompression in cases with mass effects.9) Similarly, Tsutsumi et al.3) highlighted that substantial medullary compression may present with non-specific symptoms, necessitating careful correlation between imaging and clinical findings. Postoperative improvements in spontaneous respiration and radiological decompression in our patient demonstrated the mechanical efficacy of the surgery, although full functional recovery within a short duration was limited by prolonged preoperative deficits.
The EETA is a viable option for surgical decompression of thrombosed VA aneurysms that cause brainstem compression, particularly when standard endovascular or conventional surgical interventions are insufficient. This approach enhances surgical access and visualization in the ventral brainstem, while potentially minimizing brainstem manipulation. Further investigation is warranted to better define the indications, efficacy, and safety of the management of complex thrombosed aneurysms.
Author Ryuta Saito is one of the Editorial Board members of the Journal. This author was not involved in the peer-review or decision-making process for this paper.
We thank all staff members of the Department of Neurosurgery at Nagoya University for their support and for taking care of the patient in this study. No funding was received for this research.
Conception and design: IGKASK and KI. Data acquisition: IGKASK, KI, and YS. Data analysis and interpretation: KI, KT, and YS. Drafting of the article: IGKASK and KI. Critical revision of the article: IGKASK, KI, and KT. Review of the submitted version of the manuscript: All authors. IGKASK approved the final version of the manuscript on behalf of all authors. Administrative/technical support: KI, MN, YN, and EO. Study supervision: KI, KT, IWN, TGBM, and RS.
All authors have no conflict of interest.
The data supporting the findings of this study are available from the corresponding author, KI, upon reasonable request.
