Visualization of Distal Vessels during Mechanical Thrombectomy using Aortic Cone-beam Computed Tomography

Abstract

During mechanical thrombectomy, manipulation under the limited visibility of vessels distal to the occlusion site could result in vessel perforation. Standard imaging techniques such as magnetic resonance imaging and digital subtraction angiography often fail to reveal detailed structures beyond the occlusion site. Aortic cone-beam computed tomography, that is cone-beam computed tomography performed using an intraarterial injection of contrast medium from the ascending aorta, has the potential to overcome this limitation by providing a comprehensive view of intracranial vessels. This study investigated whether aortic cone-beam computed tomography could better visualize distal vessels beyond occlusion than other modalities. This retrospective cohort study included patients with acute ischemic stroke who underwent mechanical thrombectomy during 2020-2024. The visibility of distal vessels was scored on a 5-point scale according to the visualized segment of vessels beyond the occlusion site (1 indicates excellent, and 5 indicates poor) and then compared between aortic cone-beam computed tomography and conventional imaging techniques, including digital subtraction angiography and magnetic resonance imaging, using the Wilcoxon signed-rank test. This study included 100 patients. Aortic cone-beam computed tomography resulted in a median visibility score of 2, which was better than those obtained by digital subtraction angiography (2 vs 3, p < 0.001) and magnetic resonance imaging (2 vs 5, p < 0.001). Aortic cone-beam computed tomography improves the visibility of distal vessels beyond occlusion than magnetic resonance imaging and digital subtraction angiography. This may help in safe procedures during mechanical thrombectomy.

Introduction

One of the unsolved problems during mechanical thrombectomy is lack of visualization of the vessel distal to the occlusion. In particular, crossing the clot without a clear roadmap¹⁾ increases the risk of perforation and fatal hemorrhage.^2,3)

Although magnetic resonance (MR) imaging (MRI) and computed tomography (CT) angiography (CTA) are commonly used for diagnosis and decision-making for mechanical thrombectomy, they often provide limited information regarding vessels beyond the occlusion site.⁴⁾ Aortic cone-beam CT (Ao-CBCT), that is―CBCT performed using an intraarterial injection of contrast medium from the ascending aorta―has emerged as a potential solution to the abovementioned issue.⁴⁾ Ao-CBCT can visualize all the intracranial arteries together, allowing visualization beyond the occlusion site due to the contrast medium from collaterals.⁴⁾ However, no study has yet directly compared Ao-CBCT with other modalities.

Therefore, we conducted this study to investigate whether Ao-CBCT could provide better visualization of vessels beyond the occlusion and discuss additional advantages of Ao-CBCT in comparison with other modalities.

Materials and Methods

Study design and setting

This retrospective cohort study was conducted at a single comprehensive stroke center that serves as a major referral center for stroke management in the region. In our hospital, MRI and MR angiography are the standard first-line studies to evaluate stroke patients, and mechanical thrombectomy is performed according to the Japanese guidelines.^5,6) As a standardized procedure, the transfemoral approach was used, and Ao-CBCT was performed before navigating the guiding catheter to the internal carotid artery (ICA) or vertebral artery. Consecutive cases of adult patients, aged over 18 years, who underwent mechanical thrombectomy for acute ischemic stroke at our hospital between January 2020 and October 2024 were included in this study. Eligibility for inclusion required a diagnosis of acute ischemic stroke caused by large or medium vessel occlusion (ICA, middle cerebral artery [M1 or M2 depending on the occluded site], and basilar artery [BA]), confirmed by the initial MRI upon hospital admission. Patients who did not undergo Ao-CBCT or patients who achieved spontaneous recanalization before Ao-CBCT were excluded.

Data collection and outcomes

Data were collected from the medical records and hospital stroke registry, which contains comprehensive details of each stroke management. The primary outcome was the quality of visualization of vessels beyond the occlusion site using MRI, digital subtraction angiography (DSA), and Ao-CBCT, which was rated according to a 5-point scale (visibility score, 1 indicates excellent, and 5 indicates poor) as described in previous studies,^4,7) where 1 indicates visualization of the distal portion of the occluded segment (e.g., for M1 occlusion, the distal portion of M1 was reconstituted), 2 represents visualization of the proximal portion of the adjacent segment (e.g., for M1 occlusion, the proximal portion of M2 was reconstituted), 3 refers to visualization of the distal portion of the adjacent segment (e.g., for M1 occlusion, the distal portion of M2 was reconstituted), 4 indicates visualization up to 2 segments distal (e.g., for M1 occlusion, M3 was reconstituted), and 5 represents little or no visualization of the distal territory (Supplementary Figure 1a-d).^4,7)

Initial ipsilateral intracranial arteriography (common carotid arteriography, internal carotid arteriography, or vertebral arteriography) was used as the data of DSA. Regarding MRI, maximum-intensity-projection images from 1.5 or 3 Tesla time-of-flight MR angiography was used. The evaluation was conducted independently by 2 certified neurointerventionalists, interobserver agreement was assessed using Kappa statistics, and discrepancies were resolved by consensus.

Other variables extracted included age, sex, pre-stroke modified Rankin Scale, initial National Institutes of Health Stroke Scale (NIHSS) score, initial DWI-ASPECTS score (Diffusion-Weighted Imaging Alberta Stroke Programme Early CT score) for ICA, M1, or M2 occlusion, initial posterior circulation acute stroke prognosis early CT scores for BA occlusion, occluded vessels, and stroke etiology. Treatment details included the anesthesia method, recombinant tissue plasminogen activator, procedure for reperfusion, number of passes, time to reperfusion, modified thrombolysis in cerebral infarction (TICI) grade after treatment, and complication rates. Successful reperfusion was defined as modified TICI grade of 2b, 2c, or 3.

Imaging techniques of Ao-CBCT

Ao-CBCT procedures were conducted using a Philips imaging system (Philips, Amsterdam). The imaging conditions were standardized to ensure consistent quality across all examinations. The following specific conditions were applied for Ao-CBCT imaging: the 5-Fr inner catheter was positioned in the ascending aorta to optimize the visualization of the bilateral vertebral arteries and ICAs. Nonionic contrast medium with an iodine concentration of 300 mg/m was used. The total volume of the contrast medium was (heart rate per min) multiplied by 0.5 mL, around 40 mL in total, which was injected in 20 secs. The x-ray delay time was 10 secs, and scan time was 20 secs; therefore, the images were acquired in 30 secs. The reconstruction of the images was performed using the Interventional Workspot software (Philips, Amsterdam), which allowed a detailed evaluation of the vascular structures. The reconstruction settings were 140% cube size, 256 × 256 × 256 as resolution, and “stent” as smoothness.

Statistical analysis

Baseline characteristics of the cohort were described as number (percentage), median (interquartile range [IQR]), and mean (standard deviation) unless otherwise specified. For continuous variables, the normal distribution was checked using the graphic distribution and the Shapiro-Wilk test. Statistical analysis to compare the baseline characteristics was not necessary because of the nature of the paired test. The Wilcoxon signed-rank test was used to evaluate the differences in visibility scores between MRI and Ao-CBCT, and between DSA and Ao-CBCT. Furthermore, the good visibility (the visibility score 1 or 2) and the poor visibility (the visibility score 3-5) was compared using Fisher exact test or chi-square test. In the subgroup analysis, the cohort was divided into 4 groups according to the occluded site (ICA, M1, M2, and BA), and then the visibility scores were compared. The missing data were deleted from analyses, but the numbers of missing data were recorded. Analyses were conducted using the SPSS software (version 28, IBM Corp., Chicago, IL). A p-value of less than 0.05 was considered statistically significant, and the Bonferroni correction was used to adjust p-values when performing multiple statistical tests.

Ethics

This study was approved by the institutional ethics committee (ethical committee of Toranomon Hospital, No. 2644) and conducted according to the guidelines of the Declaration of Helsinki. Clinical data were anonymized and retrospectively collected. The reporting guidelines of Strengthening the Reporting of Observational Studies in Epidemiology were followed. Written informed consent was waived because we used clinical information obtained in routine clinical practice.

Results

Baseline characteristics of the cohort

We examined 118 patients who underwent mechanical thrombectomy for acute ischemic stroke, of whom 16 patients were excluded due to the lack of data on Ao-CBCT, primarily because of the operator's unintended skip of this study, and 2 patients were excluded because reperfusion was achieved before Ao-CBCT. Finally, the remaining 100 patients were included in the analysis.

The median age of the patients was 77 years (IQR 68-85), and 59% were men. Cardioembolic stroke accounted for 70% of the cases, whereas atherosclerosis and other embolic sources accounted for 15%, respectively. The median NIHSS score was 17 upon presentation, and their median initial DWI-ASPECT score was 7. The occluded sites were ICA in 22%, M1 in 32%, M2 in 36%, and BA in 10%. Intravenous recombinant tissue plasminogen activator was given in 71%, and the combined technique was performed in 68%, followed by a direct aspiration first-pass technique performed in 25% of the cases. (Table 1).

Table 1.

Baseline Characteristics of the Cohort

		Entire cohort (n = 100)	Missing data (n)
Age, years		77 (68-85)	0
Sex	Male	59 (59)	0
Pre-stroke mRS		0 (0-1)	0
Initial NIHSS score		17 (9-23)	1
Initial DWI-ASPECTS		7 (6-9)	0
Occluded vessel	ICA	36 (36)	0
	M1	32 (32)	0
	M2	22 (22)	0
	BA	10 (10)	0
Tandem lesion		6 (6)	0
Cause of stroke	Cardioembolic	70 (70)	0
	Atherosclerosis	15 (15)	0
	Others	15 (15)	0
rtPA		71 (71)	0
Anesthesia	Local anesthesia	97 (97)	0
First procedure	Stent retriever	5 (5)	0
	A direct aspiration first pass technique	25 (25)	0
	Combined technique	68 (68)	0
	Percutaneous transluminal angioplasty	2 (2)	0

Data are expressed as number (percentage), median (interquartile range), and mean (standard deviation) unless otherwise specified.

BA: basilar artery; DWI-ASPECTS: diffusion-weighted imaging Alberta stroke program early computed tomography score; ICA: internal carotid artery; M: middle cerebral artery; mRS: modified Rankin Scale; NIHSS: National Institutes of Health Stroke Scale; rtPA: recombinant tissue-type plasminogen activator

Visualization of distal vessels

The median visibility scores were 2 (IQR 1-2), 3 (IQR 2-5), and 5 (IQR 5-5) for Ao-CBCT (n = 100), DSA (n = 99), and MRI (n = 95), respectively. Ao-CBCT yielded significantly better visualization than DSA (p < 0.001) and MRI (p < 0.001) (Fig. 1). The good visibility (score 1-2) was achieved more often with Ao-CBCT than DSA (77% vs 45%, p < 0.01, odds ratio 13.3) and MRI (77% vs 21%, p = 0.38, odds ratio 7.7). Interrater agreement of the good visibility was sufficient in Ao-CBCT (κ = 0.70), moderate in DSA (κ = 0.60), and particularly high in MRI (κ = 0.94).

Fig. 1.

Visibility score of each imaging modality.

Ao-CBCT has better visibility scores than DSA (p < 0.001*) and MRI (p < 0.001**).

Ao-CBCT: Aortic cone-beam computed tomography; DSA: digital subtraction angiography; MRI: magnetic resonance imaging

The advantage of Ao-CBCT was consistent across occlusion sites. In ICA occlusions, its median visibility score was 3, compared with 5 on both DSA (p < 0.001) and MRI (p = 0.001). For M1 occlusions, Ao-CBCT and DSA each had a median score of 2, whereas MRI remained at 5 (p < 0.001 vs Ao-CBCT). In M2 occlusions, Ao-CBCT achieved a median score of 1, outperforming DSA at 2 (p =0.005) and MRI at 5 (p <.001). A similar pattern emerged in BA occlusions: Ao-CBCT scored 1, DSA 2 (p = 0.004), and MRI 5 (p = 0.068) (Fig. 2).

Fig. 2.

Visibility score among the occluded sites.

In ICA and M2 occlusion, Ao-CBCT has better visibility score than DSA or MRI. DSA in M1 occlusion and MRI in BA occlusion shows better visibility scores than those in other vessels. However, Ao-CBCT demonstrates the best visibility scores even in M1 and BA occlusion.

Ao-CBCT: Aortic cone-beam computed tomography; BA: basilar artery; DSA: digital subtraction angiography; ICA: internal carotid artery; M: middle cerebral artery; MRI: magnetic resonance imaging

The median pass procedure was one time, and the median time from symptom onset to reperfusion was 294 mins. Successful reperfusion was achieved in 87% of cases, and vessel perforation was not detected in any cases (0%) (Table 2).

Table 2.

Treatment Outcomes

	Entire cohort (n = 100)	Missing data (n)
Pass	1 (1.0-2.7)	0
Onset to reperfusion (min)	294 (202-525)	10
Successful reperfusion	87 (87)	0
Vessel perforation	0 (0)	0

Data are expressed as number (percentage), median (interquartile range), and mean (standard deviation) unless otherwise specified.

Discussion

This study demonstrated that Ao-CBCT visualizes the vessels beyond the occlusion more clearly than conventional imaging techniques such as DSA and MRI.

The reliability and good quality of Ao-CBCT in visualizing distal vessels have been reported in the previous case series.⁴⁾ To our knowledge, our study is the first to compare Ao-CBCT head-to-head with other conventional imaging techniques to evaluate its effectiveness. The strengths of our study are the standardized protocol of Ao-CBCT and the consecutive and comprehensive database concerning patients with ischemic stroke in our hospital, allowing a detailed evaluation of the treatment procedure and outcome. The previous study also reported a descriptive comparison of Ao-CBCT with other techniques such as timing-invariant CTA and intravenously enhanced CBCT.⁴⁾ To summarize, Ao-CBCT requires a lower volume of contrast agent and a shorter time than intravenously enhanced CBCT and CTA and also yields sufficient information.

Better visualization of vessels beyond the occlusion using Ao-CBCT

The most powerful strength of Ao-CBCT, beyond the occlusion, is the simultaneous visualization of all the intracranial arteries (bilateral carotid arteries and vertebral arteries). DSA from the proximal site of occlusion could reveal the distal portion only if abundant leptomeningeal anastomosis exists. Moreover, a recent study reported the usefulness of CBCT for visualizing the distal portion of the occluded site in medium vessel occlusion caused by the backflow from the leptomeningeal anastomosis.¹⁾ In contrast, Ao-CBCT could visualize the collateral flow not only through the leptomeningeal anastomosis but also through the anterior communicating artery and posterior communicating artery and the extracranial-intracranial anastomosis. Hence, it could visualize the distal portion more certainly than DSA or conventional CBCT, especially in cases of BA occlusion, ICA occlusion, or middle cerebral artery occlusion with hypoplasia of the ipsilateral anterior cerebral artery.

Advantages of Ao-CBCT

Better visibility of distal vessels provides numerous advantages in mechanical thrombectomy. First, with Ao-CBCT provides a clear view of the vessel beyond the occlusion, resulting in avoidance of vessel perforation especially in cases with tortuous vessels or the presence of an aneurysm at the distal site (Fig. 3). Several studies shows that vessel perforations occur in 1%-4.9% of mechanical thrombectomy,^2,3,8) resulting in lethal conditions with 30%-56% mortality.^2,3) In our cohort, no perforations occurred, possibly supporting the usefulness of Ao-CBCT.

Fig. 3.

Representative cases of Ao-CBCT.

In a case of left M1 occlusion (thick arrow in Fig. 3a/b/c), Ao-CBCT could reveal the proximal part of M2 (arrow heads in Fig. 3b/c), whereas MIP from MR angiography could not. The other case is the right M1 occlusion (thick arrow in Fig. 3d). Ao-CBCT could be used as a 3D roadmap (Fig. 3e) and assists procedures of lesion-cross and stent deployment.

3D: 3-dimensional; Ao-CBCT: Aortic cone-beam computed tomography; M: middle cerebral artery; MIP: Maximum intensity projection; MR: magnetic resonance

Second, Ao-CBCT allows for better evaluation of the clot length, vessels caliber, or morphology; thereby facilitating the selection of appropriately sized devices for thrombectomy. Insufficient length of stent retriever is associated with embolization in new territories,⁹⁾ and oversized stent retrievers may carry high risks of developing hemorrhagic complications. Furthermore, some recent studies have mentioned the importance of a high catheter-to-vessel ratio of the aspiration catheter for successful reperfusion,¹⁰⁾ whereas an aspiration catheter with an extremely large ratio can cause intracranial hemorrhage,¹¹⁾ thus increasing the importance of recognizing the vessel size during thrombectomy. To summarize, Ao-CBCT may help clinicians decide the optimal devices, resulting in safer and more effective procedures.

Third, Ao-CBCT also serves as an effective tool for planning working angles during the procedure. By providing a 3D roadmap, Ao-CBCT enables the interventionalist to select the most optimal angles for lesion-cross and device deployment, which might help increase procedural efficiency and safety.

Finally, Ao-CBCT may provide advantages in the direct transfer setting, where patients are transferred from the emergency room directly to the angiosuite. Previous studies have demonstrated the effectiveness of direct transfer, such as decreasing the time to recanalization and improving patient outcomes.^12,13) The non-contrast-enhanced flat-panel CT of the brain can be used in such scenarios to decide the indication of mechanical thrombectomy but gives little vascular details. Ao-CBCT can visualize the whole vasculature immediately upon arrival, without the need for additional diagnostic imaging such as CTA, and may shorten the total time to reperfusion.

Disadvantage of Ao-CBCT

Ao-CBCT requires an extra iodinated contrast. Although this volume (around 40 mL) is acceptable for most patients, it deserves attention in those with advanced chronic kidney disease. The scan also adds a short delay, especially while operators are learning the technique. In practice, the effect on workflow is small because acquisition lasts only 30 secs, no additional catheters are needed, and the detailed roadmap it provides can speed subsequent steps. Finally, the motion artifact can reduce the image quality. Stabilization of the head position is important, and sedation might be necessary in a restless patient.

Limitations

Our study has several limitations. First, this was a retrospective study conducted in a single center, thereby presenting a risk of selection bias. Moreover, our results were confirmed only in our cohort, indicating the need for caution in the generalization of results. Second, we did not compare CTA and Ao-CBCT because MRI is the first imaging modality for patients with stroke in our hospital. This comparison should be performed in future research because CTA, especially multiphase CTA, can also visualize vessels distal to the occluded site.¹⁴⁾ Nonetheless, Ao-CBCT still retains advantages, such as real-time 3D road-mapping and usefulness in direct-to-angiosuite pathways. Third, data regarding the delayed time due to Ao-CBCT were not collected. However, the total acquisition time was only 30 secs, which might not delay the entire procedure. The previous research using a similar protocol of CBCT reported acceptable time to the first pass and reperfusion as 4.8 and 38.8 mins, respectively.¹⁾ Finally, because every patient received Ao-CBCT, we lacked a control group, preventing a direct comparison of successful reperfusion or perforation rate. Future research through randomized, larger, multicenter, and comparative studies is necessary to further validate the findings of our study.

Conclusion

Ao-CBCT has the potential to improve the safety of mechanical thrombectomy procedures by providing better visualization of vessels distal to the occlusion. By addressing the limitations of conventional imaging modalities, Ao-CBCT could become an integral component of thrombectomy protocols, potentially contributing to safer procedures and better patient outcomes. Further studies on a larger cohort are required to confirm these benefits and explore the implications for stroke treatment protocols.

Author Contributions

Satoshi Miyamoto: Conceptualization, Methodology, Data curation, Formal analysis, Investigation, Visualization, Writing- Original draft preparation. Wataro Tsuruta: Conceptualization, Validation, and Supervision. Shuhei Egashira, Jun Isozaki, and Daiichiro Ishigami: Conceptualization, Writing-Review & Editing.

All authors read and approved the final manuscript, then gave explicit consent to submit.

Conflicts of Interest Disclosure

All authors have no conflict of interest.

Data Availability Statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Funding Statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.