Strategies and Outcomes of Endovascular Treatment for Ruptured Vertebral Artery Dissecting Aneurysms: A Single-Center Experience with 44 Consecutive Cases

Article information

Neurointervention. 2026;21(2):70-78

Publication date (electronic) : 2026 June 5

doi : https://doi.org/10.5469/neuroint.2026.00381

Haibin Chen, PhD^1,*, Qi Jia, PhD^2,*, Guosai Zhang, PhD¹, Linbo Zhao, MD^,1, Sheng Liu, MD^,1

¹Department of Interventional Radiology, The First Affiliated Hospital with Nanjing Medical University, Nanjing, China

²Neurointerventional Center, Affiliated Hospital of Nantong University, Nantong, China

Correspondence to: Sheng Liu, MD Department of Interventional Radiology, The First Affiliated Hospital with Nanjing Medical University, 300 Guangzhou Rd, Nanjing 210029, China E-mail: liusheng@njmu.edu.cn

Linbo Zhao, MD Department of Interventional Radiology, The First Affiliated Hospital with Nanjing Medical University, 300 Guangzhou Rd, Nanjing 210029, China E-mail: linberzhao@hotmail.com

*These authors contributed equally to the manuscript as first author.

Received 2026 April 6; Revised 2026 May 13; Accepted 2026 May 23.

Abstract

INTRODUCTION

Ruptured intracranial vertebral artery dissecting aneurysms (VADAs) are associated with poor prognosis. If untreated, the re-rupture rate can range from 30% to 70% and the mortality rate can reach 46.7% [1-4]. Thus, the primary goal of treatment is to prevent re-rupture, and parent artery occlusion (PAO) (including the rupture site) has been proposed as a definitive solution. However, PAO eliminates anterograde flow, thereby increasing the risk of perforator infarction or vertebrobasilar ischemia and complicating the management of acute cerebral circulation during vasospasm. In contrast, reconstructive strategies aimed at preserving antegrade flow—such as single-stent-assisted coiling (SAC), multilayer stenting, or flow diverters (FDs)—have been increasingly validated in recent studies [1-3]. Reconstructive techniques are particularly indicated in situations such as significant asymmetry in VA diameter when the dissecting aneurysm occurs in the dominant VA, or with lesions involving posterior inferior cerebellar arteries (PICAs) [5]. Limitations of these approaches include the off-label use of stents or FDs in the acute phase, a potential high risk of rebleeding, and a lack of high-level evidence guiding antithrombotic management [6].

The choice between deconstructive (PAO) and reconstructive techniques hinges on balancing the need to preserve parent vessel/perforator patency against the goal of preventing acute re-rupture. Currently, no consensus exists on risk-adaptive protocols for anatomical subtypes [7]. This study introduces our experience of endovascular treatment protocol based on VA dominance and PICA involvement, validated in 44 consecutive cases of ruptured VADAs.

MATERIALS AND METHODS

Endovascular Protocol

The endovascular treatment strategy was determined based on VA dominance and the anatomical relationship of the dissection site to PICA. For non-dominant VA lesions, PAO (a deconstructive technique) was performed, with coil embolization extended 2–5 mm proximal to the lesion. In non-dominant VA aneurysms involving the PICA, PICA-preserving techniques using small-size stent (Atlas/Leo Baby) were applied prior to occlusion (Fig. 1).

Fig. 1.

DSA showed a left ruptured VADA involving PICA (non-dominant) (A). (B) illustrates the technique for PAO with PICA stenting. The white arrow designates the marker band of the stent (Atlas) within the PICA (C). Six-month follow-up angiography demonstrated robust perfusion through the right VA with preserved patency of the left PICA (D, E). DSA, digital subtraction angiography; VADA, vertebral artery dissecting aneurysm; PICA, posterior inferior cerebellar artery; PAO, parent artery occlusion.

Reconstructive stent-based techniques were preferred for VADAs in the dominant VA. For dominant VA aneurysms with VA stenosis, single laser-cut stents (Neuroform/Enterprise) with SAC were used (Fig. 2). Fig. 3 illustrates the management of a dominant-side VADA involving PICA. When dense embolization was challenging with laser-cut stents, the low-profile visualized intraluminal support (LVIS) stent was deployed using the “lantern technique” to reconstruct the aneurysm neck and achieve optimal occlusion. For all other cases, overlapping SAC was prioritized, combing laser-cut stents for scaffolding and braided stents (LVIS) for flow diversion (Fig. 4). All cases during the study period were treated according to the above-mentioned treatment strategy.

Fig. 2.

DSA identified a VADA with distal VA stenosis (A). Single SAC (Enterprise) was performed successfully (B). The DSA follow-up revealed enlargement of the VADA (C). It was decided to release an LVIS within the original Enterprise and increase the number of coils (D, E). Six months later, the aneurysm disappeared, and the blood flow of the parent artery was unobstructed (F). DSA, digital subtraction angiography; VADA, vertebral artery dissecting aneurysm; SAC, stent-assisted coiling; LVIS, low-profile visualized intraluminal support.

Fig. 3.

Angiography revealed a right VADA adjacent to the origin of the PICA with distal arterial stenosis (A). Single SAC was performed to treat the aneurysm while preserving PICA patency (B). At 6-month angiographic follow-up, DSA confirmed preserved PICA flow (C) and complete aneurysm occlusion (D). VADA, vertebral artery dissecting aneurysm; PICA, posterior inferior cerebellar artery; SAC, stent-assisted coiling; DSA, digital subtraction angiography.

Fig. 4.

A patient was found to have a VADA located on a dominant VA (A). So, overlapping SAC was performed for treatment (B). The position of the postprocedural stents and coils was satisfactory (C). At 6-month follow-up, the aneurysm had been completely occluded (D). VADA, vertebral artery dissecting aneurysm; SAC, stent-assisted coiling.

Antiplatelet regimens included intraprocedural tirofiban and postoperative dual therapy. Tirofiban was administered intravenously as a loading dose (5 μg/kg) immediately after stent placement, followed by a maintenance dose of 0.1 μg/kg/min for 24 hours. All patients received dual antiplatelet therapy (100 mg of aspirin and 75 mg of clopidogrel) for at least 3 months, followed by monotherapy with 100 mg aspirin for at least 1 year.

RESULTS

Baseline Characteristics

The cohort included 44 patients with a mean age of 56.8±12.1 years and a male predominance (52.3%, n=23). Hypertension was the most common risk factor (47.7%, n=21), followed by diabetes (22.7%, n=10), smoking (13.6%, n=6), and alcohol abuse (11.4%, n=5). Eleven patients (25.0%) presented with an initial poor Hunt–Hess grade (≥IV), and 24 (54.5%) had a Fisher grade ≥III. The mean maximal aneurysm size was 7.1±2.9 mm. Anatomically, 25 aneurysms (56.8%) were located in the dominant VA, and 9 (20.5%) involved the PICA. Baseline characteristics are presented in Table 1.

Table 1.

Characteristics of patients and ruptured VADAs (n=44)

Clinical and Angiographic Outcomes

Favorable 3-month clinical outcomes were achieved in 30 patients (68.2%). No baseline characteristics differed significantly between the favorable and unfavorable outcome groups (all P>0.05; Table 2). Ischemic events were associated with poor clinical outcomes, particularly post-procedural ischemic events (P=0.012). Patients with an initial poor initial Hunt–Hess grades (≥IV) had significantly poorer outcomes (P=0.025).

Table 2.

Characteristics of VADAs patients with different clinical outcomes

Angiographic follow-up was performed for 34 aneurysms (6 deceased, 4 declined), with complete occlusion achieved in 32 (94.1%). Two patients with increased aneurysm volume underwent retreatment: one received an additional LVIS stent within the existing Enterprise stent followed by dense coiling, and the other patient (initially treated with overlapping SAC) underwent FD placement.

Risk Factors for Unfavorable 3-Month Outcomes

Univariate analysis identified initial poor Hunt–Hess grade (≥IV) (odds ratio [OR] 6.50, 95% confidence interval [CI] 1.47–28.70; P=0.014) and diabetes (OR 4.87, 95% CI 1.10–21.69; P=0.038) as significantly predictors of unfavorable outcomes. Multivariate regression confirmed initial poor Hunt–Hess grade (≥IV) as an independent risk factor (OR 6.35, 95% CI 1.24–32.59; P=0.027) (Table 3).

Table 3.

Risk factors for unfavorable clinical outcomes after endovascular treatment

DISCUSSION

This study evaluated 44 consecutive patients with ruptured VADAs, with treatment stratified by VA dominance: SAC for dominant VAs (25 cases: 12 single SAC, 13 overlapping SAC) and PAO for non-dominant VAs (19 cases). Procedure-related complications occurred in 6 SAC cases (24.0%), with no complications in the PAO group. Notably, 2 patients with non-dominant VAs and PICA involvement underwent complete VA occlusion with low-profile stents (Atlas/Leo Baby) to preserve PICA patency, with no procedural complications or recurrence. At our institution, low-profile stents are deployed to preserve larger-caliber PICAs when involved, and embolization is confined to 2–5 mm proximal to the VADA to minimize perforator compromise—technical refinements that likely contributed to the favorable outcomes in the PAO cohort.

PAO is effective for preventing acute re-rupture but is associated ischemic complications in previous studies [1,9,10]. Lee et al. [11] reported procedure-related ischemic events in 25.0% (6/24) of PAO-treated ruptured VADAs and an 8.3% (2/24) recurrence rate. Madaelil et al. [12] noted cerebral infarction in 20.0% (2/10) of PAO cases, including 1 patient with significant disability due to PICA territory infarction, likely secondary to vasospasm or thromboembolism. PICA involvement is a known risk factor for poor outcomes; while PICA bypass was previously favored for cases where PAO was contraindicated, advances in interventional techniques have expanded options for VADAs involving the PICA [1,7,13,14]. Chen et al. [15] reported 7 cases of stenting from the distal VA to the PICA in PICA-involved VADAs, with 1 patient (14.3%) experiencing a small lateral brainstem infarction. Cho et al. [16] described 3 patients with ruptured VADAs who underwent PAO after PICA stenting, with 1 case of cerebellar infarction. Lee et al. [11] documented 3 procedural ischemic events among 4 fusiform VADAs treated with PICA stenting, with 1 recurrence during follow-up. In contrast, our 19 PAO-treated patients (including 2 PICA preservation via small-profile stents) showed no procedural complications or recurrence.

Deconstructive treatment is challenging in dominant VAs, making reconstructive techniques necessary when vessel sacrifice is contraindicated. Single SAC is technically simpler than overlapping SAC but faces challenges with immediate occlusion rates and re-rupture complications. Zhao et al. [6] reported peri-procedural hemorrhagic events in 14.3% (3/21) of single SAC cases, comparable to our rate of 16.7% (2/12). All hemorrhagic events in our study occurred in the single SAC group, potentially due to inadequate flow diversion by a single stent to prevent re-rupture and incomplete immediate occlusion (50.0%, 6/12), which increases hemorrhagic risk. Zhao et al. [17] demonstrated that overlapping SAC significantly reduces recurrence compared to single SAC (2.1% vs. 18.0%, P=0.01). Our study was underpowered to detect a statistically significant difference (single: 8.3% vs. overlapping 7.7%). This propensity for re-rupture reminds us that antiplatelet therapy should be used with caution in cases of stent‑assisted coiling. For patients undergoing single SAC, tirofiban should be administered only after complete occlusion of the aneurysm sac has been confirmed, thereby reducing the risk of aneurysm re-rupture attributable to antiplatelet agents. For cases requiring overlapping SAC, the combination of a laser‑cut stent and a braided stent typically exert a certain flow‑diversion effect, providing sufficient protection for the aneurysm.

Ischemic complications are a key concern in overlapping SAC [18,19]. Wu et al. [20] reported a 2.9% (1/34) ischemic complication rate in ruptured vertebrobasilar dissecting aneurysms treated with overlapping SAC (e.g., LVIS-within-Enterprise), while Shi et al. [21] noted a thromboembolic rate of 9.8% (4/41) across vertebrobasilar dissecting aneurysms. Our study observed ischemic events in 23.1% (3/13) of ruptured VADAs treated with overlapping stents, a higher rate possibly due to the exclusive inclusion of ruptured VADAs (necessitating more conservative antiplatelet regimens due to SAH) and a small sample size.

Managing dominant VADAs with PICA involvement is clinical challenge, as continuous perfusion from the PICA into unprotected residual dissecting segments promote recurrence [22]. Complete embolization is inherently limited in these cases, regardless of stent technique. Cho et al. [16] reported recurrence in 25.0% (3/12) of PICA-involved VADAs treated with SAC (1/4 overlapping SAC, 2/8 single SAC). In our cohort, targeted coiling sparing the PICA orifice was to preserve PICA, as shown in Fig. 3, with only 1 recurrence (11.1%, 1/9) during follow-up. Bypass surgery (e.g., VA-PICA, PICA-PICA, or occipital artery-PICA bypass) may yield favorable outcomes for this subtype by enabling definitive aneurysm management while preserving flow in the dominant VA and PICA [23].

Data on FDs for ruptured VADAs are limited and mixed. Catapano et al. [24] reported 2 acute rebleeding events and 3 retreated among 7 patients. Chan et al. [25] noted minor procedural complications in 2 of 6 patients, with complete occlusion in all cases at follow-up. Duangprasert et al. [26] demonstrated favorable outcomes in 13 cases, with 61.5% achieving complete occlusion within 2 weeks and 92.3% good clinical outcomes, with no procedural complications. However, FD use in ruptured VADAs remains controversial due to challenges in acute-phase antithrombotic management and rebleeding risk, leading us to exclude FDs from our protocol. Off-label FD use for ruptured VADAs requires further investigation in large multicenter studies.

Unfavorable outcome for ruptured VADAs range from 40.0% to 53.8% in previous reports [11,24]. Our study identified initial poor Hunt–Hess grading (≥IV) as an independent risk factor for unfavorable outcomes, consistent with prior research [9,11,27]. A poor initial Hunt–Hess grade indicates severe neurological impairment, likely due to larger SAH volume, increased intracranial pressure, and disrupted cerebrovascular physiology, rendering patients more vulnerable to deterioration during treatment and recovery.

Our study has several limitations. Compared with some previous reports, we did not employ newer endovascular devices such as FDs or other innovative treatment strategies; thus, no novel technical insight is provided. Nevertheless, we believe that our findings still offer valuable and reliable evidence. Our study represents a relatively homogeneous cohort of patients with ruptured VADAs, and the sample size (44 consecutive cases) is larger than that of most prior single‑center series. More importantly, we strictly adhered to a predefined treatment protocol throughout the entire study period without any mid‑course adjustments. This consistency in treatment allocation enhances the internal validity of our results. Therefore, while lacking technological novelty, the robustness of our outcomes is supported by the larger, more uniform patient cohort and the unwavering application of a consistent treatment strategy. This study has selection bias due to the strict adherence to the treatment strategy selection among different treatment groups. Therefore, it is impossible to conduct a causal analysis between the treatment methods and clinical prognosis. A prospective multi-center study is supposed to be conducted for verification. Subgroup analyses (e.g., FDs) are limited by small sample sizes. Long-term data on occlusion durability, vessel patency, and neurological outcomes are lacking, which are critical for assessing delayed complications like in-stent stenosis. While our PICA preservation technique using low-profile stents shows promise, long-term efficacy requires confirmation.

CONCLUSION

Endovascular management of ruptured VADAs is safe and effective when stratified by anatomy: PAO for non-dominant VA lesions and SAC for dominant VA aneurysms. PICA preservation using small-size stents appears safe and effective for PICA-involved aneurysms but requires further study. An initial poor Hunt–Hess grade (≥IV) is an independent risk factor for unfavorable clinical outcomes.

Article information Continued