Basilar artery aneurysms (BAAs), a distinct and complex subtype of intracranial aneurysms, account for approximately 7%–8% of all intracranial aneurysms and nearly 50% of aneurysms within the posterior circulation (1, 2). Among these, basilar trunk aneurysms represent 0.95%–2.1% of all intracranial aneurysms (3, 4). Typically located in a deep-seated area anterior to the brainstem and adjacent to critical perforating arteries and major branches—such as the superior cerebellar artery (SCA), anterior inferior cerebellar artery (AICA), and posterior cerebral artery (PCA)—the complex anatomy of BAAs presents considerable therapeutic challenges. Given their high morbidity and mortality rates, prompt and effective management is crucial for improving patient outcomes and minimizing the risks of disability and death (5, 6).

Traditional microsurgical approaches are limited by the deep and compact location of the basilar artery and its proximity to the brainstem, resulting in substantial exposure risks and a high likelihood of neural injury, making them less than ideal. Endovascular coil embolization, which has been considered the “gold standard” for posterior circulation aneurysms (7), is associated with a higher risk of recurrence and complications, especially in large aneurysms (8, 9). With advancements in endovascular technology, flow diverters (FDs) with fine mesh designs have emerged as a minimally invasive alternative, achieving aneurysm occlusion by modulating local hemodynamics, promoting endothelialization of the aneurysm neck, and inducing thrombosis, as well as straightening the parent artery (10, 11). The Pipeline Embolization Device (PED; Medtronic, USA), the first FD approved for clinical use, currently has the most robust clinical and laboratory evidence (12). Composed of cobalt-chromium (75%) and platinum (25%), the PED offers improved radiopacity due to higher stiffness, though this rigidity can hinder complete deployment in tortuous vascular segments (13). The Tubridge flow diverter (TFD; MicroPort, China) is a relatively newer device comprised of a self-expanding, braided nitinol stent with flared ends, exhibiting favorable outcomes in the treatment of unruptured intracranial aneurysms (UIAs) and offering advantages of superelasticity and shape memory (14, 15). Compared to PED, the nitinol design of TFD provides greater flexibility, allowing better trackability, though the lower radial force may predispose the device to migration during microcatheter or microwire manipulation (16, 17). The reported aneurysm occlusion rate for FDs is about 90.9%, superior to traditional coil embolization techniques (18). However, the unique anatomical and physiological characteristics of BAAs raise specific challenges for FD application. Coverage of the basilar artery and its critical perforators can compromise blood supply, increasing the risk of ischemic stroke and new neurological deficits. Additionally, the efficacy of FDs in treating large or fusiform aneurysms remains controversial, with some studies reporting lower complete occlusion rates compared to smaller aneurysms. Factors such as patient-specific anatomy, aneurysm morphology, antiplatelet regimens, and intraprocedural technical aspects significantly affect both safety and outcomes.

Given these considerations, we conducted a comprehensive retrospective analysis of consecutive patients with BAAs treated with FD devices at our center. This study aims to systematically evaluate the procedural success, aneurysm occlusion rates, complications, and functional outcomes associated with FD treatment, thereby providing robust evidence to guide individualized management of basilar artery aneurysms and to inform future strategies for device and technique optimization in clinical practice.

Between January 2020 and September 2024, a total of 39 patients with basilar artery aneurysms were treated with flow-diverting stents at our neurointerventional center. Based on the predefined criteria, 9 patients were excluded, resulting in a final study cohort of 30 patients eligible for analysis. The detailed process of patient selection is illustrated in Figure 1. Thirty patients (median age group 65–69 years; 20 men, 10 women) harboring basilar artery aneurysms were included (Table 1). The mean aneurysm diameter was 10.6 ± 4.9 mm. Twenty-two patients (73.3%) were over 60 years. Presentation was dizziness in 7 (23.3%), headache in 2 (6.7%), brainstem compression in 3 (10%), and incidental in 6 (20%). Three patients (10%) had recurrent aneurysms after prior coiling. At presentation, mRS was 0 in 22 (73.3%), 1 in 5 (16.7%), and ≥ 2 in 3 (10%). Aneurysm location was the basilar trunk in 21 (70%) and the basilar tip in 9 (30%). Patient demographics, aneurysm characteristics, and treatment details are summarized in Table 2, with statistical comparisons provided.

Technical success was 100%; all flow diverters (FDs) fully covered the aneurysm neck. Eighteen patients (60.0%) received a Pipeline Flex (PED), and 12 patients (40.0%) received a Tubridge (TFD); three cases required two TFDs, and one required two PEDs. Adjunctive coiling was performed in four cases. Deployment was uneventful in all procedures. In total, 36 major branches were jailed: 17 AICAs, 3 SCAs, and 16 PCAs. Four patients (13.3%) experienced transient ischemic symptoms (three with unilateral limb numbness; one with numbness and diplopia), all of which resolved completely with vasospasm therapy before discharge. No acute branch occlusions were observed. Radiologic follow-up was available in 28 patients with DSA and in two patients with MRA or CTA. The mean duration of angiographic follow-up was 12.9 months (range, 8–20 months). Representative cases are shown in Figures 2–4. Factors associated with complete occlusion, complications, and clinical outcomes were systematically analyzed (Table 3). In univariate analysis, larger aneurysm size was significantly associated with lower odds of complete occlusion (OKM grade C/D; OR 0.47, 95% CI 0.14–0.77, p = 0.042). Multivariate Firth regression, which included aneurysm size, age, and adjunctive coiling, confirmed aneurysm size as an independent predictor of complete occlusion (OR 0.62, 95% CI 0.29–0.85, p < 0.001). For TIA complications, no significant associations were identified with aneurysm size (p = 0.2), age (p > 0.9), or other factors. Regarding favorable clinical outcome (mRS 0–1), larger aneurysm size showed a trend toward significance (OR 0.67, 95% CI 0.38–0.94, p = 0.055), while other factors including age, sex, aneurysm location, device type, and adjunctive coiling demonstrated no significant associations. At last follow-up, complete or near-complete aneurysm occlusion (CO or NC) was achieved in 16 of 18 PED-treated patients (88.9%) and in 10 of 12 TFD-treated patients (83.3%), resulting in an overall rate of 86.7% (26/30). Partial occlusion (PO) was seen in two PED cases (11.1%) and two TFD cases (16.7%). The difference in complete/near-complete occlusion rates between the two devices was compared using Fisher’s exact test due to the small sample size. The analysis revealed no statistically significant difference (p = 1.000). It is important to note that this high p-value likely reflects the limited statistical power of the study to detect differences given the small cohort sizes, particularly for the TFD group (n = 12), rather than definitive evidence of equivalence. At a mean clinical follow-up of 12.9 months, 22 patients (73.3%) had an mRS of 0, 6 patients (20.0%) had an mRS of 1, and 2 patients (6.7%) had an mRS of 2 at last assessment. Angiographic follow-up revealed occlusion of three jailed branches (one AICA and two PCAs, 10.0%) in three patients (Cases 10, 13, and 23).

Univariate analysis identified aneurysm size as significantly associated with complete occlusion (OKM Grade C/D) (OR = 0.47, 95% CI: 0.14–0.77, p = 0.042). Age, sex, aneurysm location, device type (PED vs. TFD), and the use of adjunctive coiling did not demonstrate significant associations in the univariate model (all p > 0.05). Multivariate analysis using Firth logistic regression confirmed that smaller aneurysm size was an independent predictor of complete occlusion (adjusted OR = 0.62, 95% CI: 0.29–0.85, p < 0.001). After adjustment, neither age (adjusted OR = 1.10, 95% CI: 0.88–1.66, p = 0.4) nor adjunctive coiling (adjusted OR = 0.85, 95% CI: 0.04–90.4, p > 0.9) showed significant association with the occlusion outcome. Four patients (13.3%) experienced transient ischemic attacks (TIA). Univariate analysis revealed no significant associations between TIA occurrence and patient age (OR = 1.01, p > 0.9), sex (OR = 0.53, p = 0.6), aneurysm size (OR = 1.15, p = 0.2), location (OR = 1.87, p = 0.6), device type (OR = 0.45, p = 0.5), or adjunctive coiling (OR = 2.56, p = 0.5). Twenty-eight patients (93.3%) achieved favorable clinical outcomes (mRS 0–1) at follow-up. Univariate analysis showed no significant associations between clinical outcome and age (OR = 0.95, p = 0.6), sex (OR = 1.80, p = 0.7), aneurysm size (OR = 0.67, p = 0.055), location (OR = 1.80, p = 0.7), device type (OR = 0.65, p = 0.8), adjunctive coiling, or TIA complications. Aneurysm size showed a trend toward significance but did not reach the statistical threshold (p = 0.055).

BAAs present a formidable challenge in neurointerventional surgery, attributable to their deep-seated location adjacent to the brainstem, their proximity to critical perforating arteries, and complex regional hemodynamics. While microsurgical clipping has been largely superseded for these lesions due to associated high morbidity and mortality, the advent of FDs has markedly transformed the therapeutic landscape. In our retrospective cohort of 30 consecutive BAA patients treated with FDs, we observed a 100% technical success rate, an overall complete or near-complete occlusion rate of 86.7% at a mean follow-up of 12.9 months, and a symptomatic complication rate of 13.3% comprising transient ischemic events. These outcomes underscore the considerable promise of FD technology for managing these complex aneurysms.

He complete or near-complete occlusion rate of 86.7% achieved in our FD-treated cohort compares favorably with historical outcomes reported for conventional endovascular techniques. As summarized in Table 4, large meta-analyses of conventional coil embolization for posterior circulation aneurysms report complete occlusion rates ranging from 70.0% to 72.9% (19–21). Stent-assisted coiling (SAC) exhibits variable efficacy, with reported rates between 59.2 and 81% (22–24). Similarly, microsurgical clipping, though highly effective in selected cases, is associated with significant risks due to the complex anatomy of the basilar artery, with reported complete occlusion rates ranging from 80% to 90% but at the cost of higher morbidity and mortality rates (21, 25). In contrast, contemporary FD series, including the present study, consistently report mid-term complete or near-complete occlusion rates between 77% and 88.3% (22, 26–30), positioning FDs as a highly effective modality. This consistent performance highlights the fundamental mechanistic advantage of FDs: parent vessel reconstruction and intra-aneurysmal flow disruption promote progressive thrombosis and endothelialization across the neck, a mechanism distinct from simply filling the aneurysm sac with coils.

Aneurysm size is a well-established factor influencing occlusion outcomes after FD treatment. A recent meta-analysis of the Surpass Evolve flow diverter demonstrated an inverse relationship between aneurysm diameter and complete occlusion rate (31), consistent with the hemodynamic principle that flow disruption is more effective in smaller aneurysms with less voluminous inflow zones. In our cohort, the mean aneurysm diameter was 10.6 mm, and the high occlusion rate likely reflects appropriate patient selection based on morphological characteristics. Our multivariate analysis further identified larger aneurysm size as an independent predictor for lower odds of complete occlusion (adjusted OR = 0.62, p < 0.001), aligning with established literature. This reinforces the importance of meticulous case selection to optimize outcomes.

In our series, both the Pipeline Flex and Tubridge devices achieved high and comparable mid-term complete or near-complete occlusion rates (88.9% vs. 83.3%, p = 1.0). However, this lack of statistical significance must be interpreted with caution due to the limited sample size, particularly the smaller Tubridge subgroup (n = 12), which precludes definitive conclusions regarding equivalence. The high p-value primarily reflects the study’s limited power to detect clinically relevant differences. Larger, prospective comparative studies are needed. The IMPACT trial, evaluating the Tubridge device in posterior circulation aneurysms (n = 200), reported a 1-year complete occlusion rate of 79% with low rates of in-stent stenosis (3.6%) and symptomatic stroke (4.5%) (32). The device’s high metal coverage and optimized pore density are designed to profoundly alter flow dynamics. Notably, all four large aneurysms (≥ 13 mm) in our Tubridge cohort achieved complete occlusion, suggesting particular utility for complex morphologies. Similarly, the SEASE study of the Surpass Evolve device (n = 305) reported a 6-month occlusion rate of 73% with a major complication rate of 2.1% (33), corroborating the efficacy of modern FD designs. The single case of partial occlusion in a previously coiled aneurysm (OKM grade B) in our Pipeline group hints at potential challenges in such scenarios, possibly due to altered hemodynamics or a pro-inflammatory milieu impeding endothelialization.

The principal risks of FD treatment in the basilar territory—perforator ischemia, thromboembolism, and in-stent stenosis—were observed in our cohort as a 13.3% rate of transient ischemic symptoms. This aligns with the broad range of FD-related complication rates reported in the literature (0.8%–17.1%) (34, 35). Several factors likely contributed to our acceptable safety profile. First, the implementation of personalized antiplatelet therapy, guided by platelet function testing, was critical. In our cohort, 20% of patients identified as clopidogrel non-responders were switched to ticagrelor, an strategy supported by contemporary evidence (32, 36) for optimizing platelet inhibition and reducing thromboembolic risk. Second, advances in procedural technique, including the use of triaxial access systems and meticulous device sizing, contributed to safe deployment. Although 10% of patients developed jailed branch occlusions [similar to the 11%–19% reported with the FRED Jr. Device (36)], no acute branch occlusions or in-stent thromboses occurred. Emerging technologies like virtual deployment platforms (37) hold promise for further optimizing device placement and predicting hemodynamic outcomes.

Our study has several limitations that warrant consideration. First, the retrospective, single-center design and modest sample size (n = 30) limit the generalizability of our findings and the power for robust subgroup analyses, particularly for comparing devices. This is partly inherent to the study population, as the use of FDs for BAAs remains selective and is not yet a first-line treatment at many centers due to the complex anatomy and perceived risks. The mean angiographic follow-up of 12.9 months is insufficient to assess long-term durability, including risks of delayed in-stent stenosis or aneurysm recanalization. Second, the single-center design inherently introduces potential selection bias. Furthermore, when comparing our outcomes with historical data from other centers or earlier time periods, one must consider the potential influences of evolving patient selection criteria, advancements in device technology, refinements in antiplatelet regimens, and improved operator experience. These temporal and institutional differences may introduce bias when attempting direct comparisons. Furthermore, the exclusion of acutely ruptured and perforator-dominant aneurysms means our results are primarily applicable to carefully selected, unruptured BAAs. Future research should prioritize prospective, multicenter registries with long-term follow-up. Investigations into hybrid devices (e.g., combining branch-sparing stents with FDs), computational fluid dynamics for pre-procedural planning, and genotype-guided antiplatelet therapy represent promising avenues for further improving the safety and efficacy of FD treatment for these challenging lesions.

This single-center experience suggests that flow diverter devices represent a promising treatment modality for carefully selected patients with basilar artery aneurysms. The primary finding of this study is the achievement of technically successful deployment and favorable mid-term complete or near-complete occlusion rates (86.7%) with an acceptable safety profile in this complex anatomic location. The observed clinical outcomes underscore the importance of meticulous patient selection and preoperative planning. However, the small sample size, retrospective design, and lack of a control group preclude definitive conclusions regarding superiority over other treatments or generalizability. These preliminary findings warrant further validation through larger, prospective, and comparative studies to establish long-term efficacy and optimize patient selection criteria.