Risk Factors for Incomplete Occlusion in Patients with Small Intracranial Aneurysms (<7 mm) after Flow-Diversion Treatment: A Multicenter Experience

The prevalence of unruptured intracranial aneurysms (UIAs) affects approximately 3–7% of adults [1]. Preventive intervention with a UIA can reduce the risk of subarachnoid hemorrhage (SAH), consequently enhancing the number of life years with a good functional prognosis [2]. Small UIAs (<7 mm) are relatively common, accounting for over 70% of all UIAs, and are increasingly detected incidentally with high-resolution cerebrovascular angiography [3]. The rupture risk is known to increase with enlargement in aneurysm size. However, several studies have reported that the risk of rupture of small UIAs can be even as high as 40% [4, 5].

Although endovascular treatment has become the mainstay of therapy in UIAs, the current embolization technique of small UIAs is encountering nonnegligible challenges (i.e., the stability of the microcatheter position, initial selection, and the risk of intraoperative perforation) [6, 7]. With advances in neurointerventional devices, there is an increasing trend for off-label use of flow diverters (FDs) for treating small, blood-blister, or distally located aneurysms, with satisfactory outcomes reported [6, 8‒11]. Due to the therapeutic concept of progressive healing, however, certain aneurysms may present incomplete occlusion (ICO) after FD implantation in long-term follow-up [12]. Thus, it is necessary to evaluate the factors affecting aneurysm occlusion.

Few centers have collected a sufficient sample population to perform a robust analysis of treatment outcomes for small UIAs [6, 8]. To address this concern, we conducted a multicenter study to determine potential triggers for ICO after FD implantation.

We performed a retrospective analysis of a prospectively maintained database at five comprehensive hospitals in China, all of which conducted over 200 endovascular treatments for intracranial aneurysms annually. We reviewed consecutive patients who underwent first-ever FD implantation between September 2018 and September 2022. This study was approved by Local Ethics Committees, and each participant provided written informed consent to participate in the study.

Adult patients with small UIAs who received FD treatment with either a pipeline embolization device (Medtronic, Minneapolis, USA) or Tubridge flow diverter (TFD; MicroPort NeuroTech, Shanghai, China) were initially reviewed. The exclusion criteria were as follows: (i) ruptured aneurysms; (ii) aneurysms received from previous endovascular or micro-neurosurgical treatment; (iii) tandem treatment (two or multiple aneurysms covered by a single FD); (iv) poor imaging quality; (v) presence of other cerebrovascular disease (i.e., arteriovenous malformations, arteriovenous fistulas, or Moyamoya disease). Additionally, since our study outcome was aneurysm occlusion status, patients without angiographic follow-up were excluded.

The antiplatelet regime has been reported in our previously published study (online supplementary Data S1; for all online suppl. material, see https://doi.org/10.1159/000544991) [13]. The selection of FD was based on the experience and preferences of the participating centers. When aneurysms with a larger vessel diameter, surgeons preferred selecting a TFD owing to its broader vascular adaptability [13]. Conjunctive coils may be required in three conditions: (i) there was a risk of stent shortening or migration; (ii) a higher risk of delayed rupture or recurrence (rapid jet flow velocities at the neck); and (iii) aneurysms with irregular shape.

The methods of data collection are described in Data S2.

Perioperative complications include intraoperative perforation, in-stent thrombosis, postoperative subarachnoid hemorrhage, postoperative intraparenchymal hemorrhage, transient ischemic attack (a transient neurological deficit without corroborative imaging), mild ischemia (a change in the National Institutes of Health Stroke Scale (NIHSS) score of ≤4 lasting for <7 days), and severe ischemia (a change in the NIHSS score of >4 lasting for >7 days) [13]. Three neurointerventionalists with more than 15 years of experience supervised these measurements.

The detailed follow-up regimes have been reported previously [13]. Aneurysm occlusion status was categorized into two groups according to the O’Kelly-Marotta classification: complete occlusion (CO, entirely non-filling) and ICO (entry remnant or subtotal and total filling). Retreatment was considered for patients who showed ICO over 24 months. In-stent stenosis (ISS) was defined as a narrowing of the FD diameter exceeding 50% without pre-existing significant artery stenosis [14]. Clinical follow-up was conducted via face-to-face outpatient visit or telephone follow-up. Based on the mRS, clinical outcomes were categorized as favorable (0–2), morbidity (3–5), and mortality (6).

All statistical analyses were performed using R studio (version 4.2.2). The normality of continuous variables was assessed with the Kolmogorov-Smirnov test. Normally distributed variables are expressed as the mean ± standard deviation and tested using the Student’s t test. Nonnormally distributed variables are presented as medians [interquartile range (IQR)] and tested using the Wilcoxon signed rank test. Categorical variables were summarized as frequency and percentage and analyzed using the Chi-square test or Fisher’s exact test. For model development, predictive variables showing a trend (p < 0.10) from univariate analysis were entered into binary logistic regression analysis with a backward stepwise elimination approach to simplify the model. Statistical significance was set at p < 0.05.

After applying the exclusion criteria, 565 consecutive participants (mean age: 53.70 ± 11.23 years, female: 71.3%) were ultimately enrolled (Fig. 1). Hypertension was the most common cardiovascular comorbidity, presenting in 34.9% of the patients. Practically all patients had an mRS score of 0–2 at admission (98.4%), while the majority of aneurysms were located in the anterior circulation (96.6%). The median aneurysm size and neck width were 4.00 mm [IQR: 2.91–5.16] and 3.59 mm [IQR: 2.70–4.60], respectively. A total of 351 pipeline embolization devices (62.1%) and 214 TFDs (37.9%) were implanted successfully, with adjunctive coils used in only 35 cases (6.2%). Detailed cohort characteristics are presented in Table 1.

After FD implantation, a total of twelve cases (2.1%) of perioperative complications were observed (Table 1), including two instances of in-stent thromboses, two postoperative intraparenchymal hemorrhages, five transient ischemic attacks, and three mild ischemia. The two cases of in-stent thrombosis received intra-arterial tirofiban injection, followed by postoperative intravenous tirofiban infusion for 24 h and dual antiplatelet bridging therapy, without developing permanent complications.

The mean durations of angiographic and clinical follow-up were 10.64 ± 5.99 and 19.36 ± 3.88 months, respectively. Complete aneurysm occlusion was achieved in 449 (79.5%) cases, with 32 cases (5.7%) developing ISS. No aneurysm recurrence was observed, and the favorable prognosis rate was 98.6%, without mortality observed during this period.

All aneurysms were categorized into CO and ICO groups based on the occlusion status during follow-up (Table 1). The older age (p = 0.029), hypertension (p < 0.001), coronary disease (p = 0.001), smoking behavior (p = 0.018), larger aneurysm size (p < 0.001), wider neck (p = 0.005), and higher size ratio (SR) (p = 0.001) were more prevalent in the ICO group.

Predictive variables in univariate analysis (p < 0.10, online suppl. Table S1) were entered into binary logistic regression analysis with a backward stepwise elimination approach. After adjusting for candidate variables, hypertension (adjusted odds ratio [aOR] = 2.274, 95% confidence interval [CI] = 1.462–3.538, p < 0.001), coronary disease (aOR = 2.742, 95% CI = 1.148–6.552, p = 0.023), larger aneurysm size (aOR = 1.833, 95% CI = 1.425–2.356, p < 0.001), lower SR (aOR = 0.380, 95% CI = 0.166–0.869, p = 0.022), and less coil application (aOR = 0.212, 95% CI = 0.061–0.741, p = 0.015) were independently associated with ICO of small UIAs (Table 2).

The proportion of small UIAs is high and the technical challenges of using traditional coil embolization cannot be neglected. Clinically, off-label use of FDs for smaller aneurysms is already common, especially in the European population [6]. Based on the PREMIER trial, the indicator of FDs for small aneurysms received an approval extension [15]. However, limited by the small sample size, the related factors for ICO of small UIAs remain unclear. In the presented study, we analyzed the treatment outcomes of FD implantation for small aneurysms at five comprehensive hospitals. Our findings demonstrated that hypertension, coronary disease, aneurysm morphology, and adjunctive coils were independently related to the occlusion status of small UIAs after flow-diversion treatment.

Differing from the mechanism of common embolization, FDs have a higher wire density, progressively promoting aneurysm healing by leading to the adherence of inflammatory cells and endothelial cell growth [16]. Hypertension is a well-established risk factor for vascular burden. Previous studies have reported that hypertension accelerates atherosclerosis through numerous mechanisms, such as cellular migration, extracellular matrix, inflammation, and levels of infiltrating plasma macromolecules [17, 18]. Therefore, a history of hypertension may potentially weaken the capacity of endothelialization, resulting in a lower rate of CO. Considering that over half of patients with intracranial aneurysms have hypertension [19], the management of patients with aneurysms deserves our significant attention, especially those with uncontrolled hypertension. Similarly, coronary disease is well known to be associated with the atherosclerotic burden in systemic circulation [20]. In the presented study, coronary disease was one of the independently related factors for ICO of small UIAs (aOR = 2.742, p = 0.023), which may reflect endothelial dysfunction and vessel fragility in aneurysm patients with coronary disease compared with those without coronary disease.

Aneurysm size is a well-known factor affecting occlusion in endovascular treatment. Consistent with previous cohorts [21, 22], our findings demonstrated that an increase in aneurysm size is a strong predictor of ICO after flow-diversion therapy for small UIAs (aOR = 1.833, p < 0.001). Aneurysm healing is a progressive process, and larger aneurysm may require more time for the formation of an intraluminal hemostatic thrombus, potentially causing a delay in the occlusion. Moreover, larger aneurysms typically experience more complex flow within the aneurysm sac, resulting in a diminished effectiveness of the FD and consequently delaying the point of aneurysm obliteration.

Additionally, we found that less coil application was independently related to incomplete aneurysmal occlusion (aOR = 0.212, p = 0.015). Similar to our results, several studies have reported that coiling-assisted embolization is more effective in aneurysm occlusion than FD treatment alone [16, 23]. Stagnation of blood flow by coil packing may facilitate thrombosis formation in the aneurysm, leading to occlusion [24]. However, more operations in the small aneurysm sac may increase complexity, such as initial selection, coil flexibility, and the risk of intraoperative perforation for the procedure. Furthermore, excessive use of endovascular devices could lead to higher treatment costs. Thus, neurointerventionists should be prudent when pursuing coil embolization in FD treatment for small UIAs.

SR was demonstrated as a significant parameter for aneurysm rupture risk [25], and it considers not only the aneurysmal morphology itself but also the local vessel diameter. However, the association between it and aneurysm occlusion status is pendent. As SR increases, the simple, stable vortex may turn into complex patterns with multiple vortices [25], which seems to contradict our results. When conducting a subgroup analysis (online suppl. Table S2), we observed that surgeons preferred to utilize coil-assisted embolization to mitigate the risk of delayed rupture or recurrence when encountering aneurysms with a higher SR (49.2% vs. 71.4%, p = 0.018). Therefore, it is unsurprising that SR and ICO are negatively correlated in multivariate analysis. In the future, conducting a detailed investigation into the potential relationship between morphology parameters and occlusion status would be highly valuable in FD implantation for small aneurysms.

To our knowledge, this was the largest study to investigate the risk factors for ICO in small UIAs following FD implantation. However, several limitations should be noticed in our study. First, the retrospective nature limits the strength of our results. Second, despite being multicenter cohorts, differences in platelet function tests and antiplatelet regimens in patient care may introduce potential bias. Third, the inhibition of platelet aggregation after drug administration was not assessed; the relationship between platelet function inhibition and the incidence of ICO could not be clarified. Finally, the relatively short mean imaging follow-up duration (10.64 ± 5.99 months) may not be sufficient for observation. Thus, our findings need to be validated by further prospective randomized controlled trials.

We found that a history of hypertension and coronary disease, larger aneurysm size, lower SR, and less coil application were independent predictors of ICO for small UIAs after FD implantation. This may suggest that neurointerventionalists should pay more attention to blood pressure management and aneurysm morphological assessment in flow-diversion treatment for small UIAs.

This study involves human participants and was approved by the Institutional Review Board of Zhujiang Hospital (2023-KY-023–02), the First Affiliated Hospital of Chongqing Medical University (CY2023-067–01), Beijing Tiantan Hospital, Capital Medical University (KY2022-102–02), Guangdong Provincial People’s Hospital (KY-Q-2022–344-02), and the Affiliated Hospital of Southwest Medical University (KY2023219). Each participant provided written informed consent to participate in the study.

The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.

This current study was supported by the Foundation of the National Health Commission Capacity Building and Continuing Education Center (GWJJ2022100102, GWJJQ2022100104) and the National Natural Science Foundation of China (82,171,290; 82,201,427).

All authors contributed to the manuscript, satisfying the ICMJE guidelines for authorship. Chuanzhi Duan is responsible for the overall contents as guarantors. Chi Huang and Xingwei Lei designed the study. Chi Huang drafted the manuscript. Xin Tong and Zhuohua Wen analyzed and interpreted the data. Jiancheng Lin, Mengshi Huang, Chao Peng, Tao Wang, Wenxin Chen, Lele Dai, and Xin Jin collected the data. Shixing Su, Xin Zhang, and Xifeng Li contributed to the management and critically revised the study outcomes. Xin Feng, Xifeng Li, Zongduo Guo, Aihua Liu, and Chuanzhi Duan conceptualized the study, contributed to funding and study supervision, and critically revised the manuscript.

Chi Huang and Xingwei Lei contributed equally to this work.ClinicalTrials.gov (NCT06446778, Registered on May 22, 2024).

The data that support the findings of this study are not publicly available due to privacy reasons but are available from the corresponding author upon reasonable request.