Vessel wall MRI predictors of postprocedural new ischemic brain lesions after endovascular treatment for basilar artery stenosis Authors: Wanda Shi, Hao Wang, Tao Wang, Shuaiwei Guo, Long Li, Yabing Wang, Peng Gao, Yanfei Chen, Jian Chen, Fei Chen, Bin Yang, Yan Ma, Yanxin Zhao, Jichang Luo, Liqun Jiao Published: 2026 ABSTRACT BackgroundDiffusion-weighted imaging (DWI)-detected new ischemic brain lesions (NIBL) are common complications after endovascular treatment (EVT) for basilar artery stenosis (BAS), directly impacting functional outcomes in patients. This study aimed to explore the association between vulnerable plaque markers identified by high-resolution magnetic resonance imaging (HR-MRI) and NIBL, and to clarify their predictive value for postprocedural ischemic events.MethodsA total of 373 patients with BAS who underwent EVT were retrospectively enrolled in this study. Patients were classified into DWI-positive (DWI+) and DWI-negative (DWI−) groups based on the presence of NIBL on DWI within 48 hours post-procedurally. Plaque burden, distribution, remodeling type, enhancement rate, and degree of stenosis were evaluated using HR-MRI. To determine independent predictors of NIBL post-procedurally, multivariable logistic regression was applied.ResultsNIBL developed in 188 patients (50.4%), with a mean age of 61.76±7.83 years, including 84 males. The DWI+ group exhibited significantly higher rates of positive remodeling (PR) (51.1% vs. 37.8%, P=0.01) and plaque enhancement (PE) (87.2% vs. 75.7%, P=0.004), as well as a greater plaque burden (13.38±40.75 vs. 6.06±29.77, P=0.033), compared to the DWI− group. Multivariate logistic regression identified PE [odds ratio (OR) =2.09, 95% confidence interval (CI): 1.18–3.70, P=0.011] and PR (OR =1.77, 95% CI: 1.03–3.06, P=0.039) as independent predictors of NIBL.ConclusionsHR-MRI-derived vulnerable plaque markers (PE and PR) are closely associated with the occurrence of NIBL following EVT for BAS, providing imaging-based evidence for risk stratification and treatment optimization. Introduction Posterior circulation ischemic strokes are predominantly caused by basilar artery stenosis (BAS), a condition associated with high morbidity and mortality. With advancements in interventional techniques and devices, endovascular treatment (EVT) has become an alternative (1). However, EVT does not benefit all patients, and postprocedural new ischemic brain lesion (NIBL), which are often major contributors to poor clinical outcomes, can undermine its potential benefits. These lesions not only pose immediate symptom-related risks but also increase the likelihood of long-term functional decline, recurrent stroke, or mortality (2-5). In addition, compared with other arterial territories, the basilar artery (BA) supplies a highly dense functional region, where even small ischemic lesions can lead to significant functional deficits. As reported in the Stenting and Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis (SAMMPRIS) trial, the incidence of postprocedural symptomatic ischemic events in BAS patients was 20.8%, compared with 6.7% in other arterial territories (6). Moreover, research has found that the incidence of NIBL after vertebrobasilar artery stenting is significantly higher than after internal carotid artery stenting (56.6% vs. 28.9%) (7). Therefore, reducing the incidence of NIBL can improve the safety and efficacy of EVT. Currently, postprocedural risk factors for NIBL after EVT for BAS remain unidentified, and there is still no personalized guidance for BAS treatment. Identifying these predictive factors is urgently needed to optimize interventional strategies and improve patient selection. Vulnerable atherosclerotic plaques are key drivers of perioperative ischemic events, with their instability leading to plaque rupture, embolization, and subsequent infarction (8). These plaques can be characterized by features including their anatomical location, degree of enhancement, dimensions, surface area, and remodeling characteristics (9-12). Several studies have shown that plaque enhancement (PE), positive remodeling (PR), intra-plaque hemorrhage, and fibrous cap rupture are considered markers of vulnerable plaques, which are associated with stroke events (13-15). However, few studies have explored the relationship between postprocedural NIBL and vulnerable plaques in patients with severe BAS. Therefore, this study focused on the correlation between high-resolution magnetic resonance imaging (HR-MRI)-identified vulnerable plaque markers and NIBL, and evaluated the predictive ability of different combinations of risk factors across various treatment modalities. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1-2677/rc). Methods Participants This study included patients who underwent EVT for BAS at Xuanwu Hospital, Capital Medical University between January 2012 and April 2022. The inclusion and exclusion criteria were consistent with those used in previous studies from our center (16).The inclusion criteria were as follows: (I) angiographically confirmed BAS exceeding 70% on digital subtraction angiography (DSA); (II) presence of ischemia-related symptoms attributable to BAS, including dizziness, vertigo, cephalalgia, diplopia; (III) undergoing endovascular therapy; (IV) completion of preprocedural HR-MRI; and (V) acquisition of MRI scans within the perioperative period, defined as 48 hours before and after intervention. The exclusion criteria were as follows: (I) history of acute stroke due to BA occlusion; (II) concurrent endovascular intervention on other cerebral or extracranial vessels; (III) coexisting moderate to severe vertebral artery stenosis; and (IV) non-atherosclerotic etiologies of BAS, such as moyamoya disease, vasculitis, or arterial dissection. The CASSISS protocol (China Angioplasty and Stenting for Symptomatic Intracranial Severe Stenosis) was followed for all procedures (17). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study involved human participants and was approved by the Ethics Committee of Xuanwu Hospital, Capital Medical University (approval No. [2013]004). Written informed consent was provided by all participants prior to their enrollment in the study. Endovascular procedure and HR-MRI protocol This study comprehensively documented patient baseline characteristics and EVT procedural details, including primary angioplasty (PA), balloon-mounted stents (BMS), and self-expanding stents (SES). All endovascular procedures were performed by neurointerventionalists with more than 15 years of experience in EVT. All patients received a standardized antiplatelet regimen consisting of aspirin (100 mg daily) and clopidogrel (75 mg daily) starting 5 days prior to the procedure. Device selection and procedural strategies followed the standard operating procedures of our center. Balloon or stent sizes were determined according to the reference vessel diameter, with the balloon diameter generally set at 0.8 times the normal reference vessel diameter. For BMS, direct dilatation was performed; for PA and SES, pre-dilatation was routinely applied. Given the high risk of perforator occlusion in the basilar artery, post-dilatation was performed only when necessary and used sparingly. Residual stenosis was required to be <30% after the procedure. In the multivariate regression analysis, adjustments were made for lesion length and device type. Periprocedural complications were documented, with a primary focus on ischemic events; other complications were not analyzed in this study due to their low incidence and will be investigated in future research. All eligible patients were examined using a 3.0T MRI system (Verio, Siemens, Erlangen, Germany) with a standard 8-channel head coil to assess the characteristics of BAS. Stenosis evaluation was conducted through time-of-flight magnetic resonance angiography (TOF-MRA), which also guided the adjustment of the HR-MRI scanning plane to ensure it was perpendicular to the target vessel. The primary high-resolution vessel wall imaging sequences included TOF-MRA, T1-weighted fast spin-echo, and contrast-enhanced T1-weighted imaging. All post-contrast T1-weighted images were acquired at 5 minutes after intravenous administration of gadopentetate dimeglumine, and this standardized time point was strictly adhered to in all patients. Comprehensive technical specifications of the imaging protocols have been systematically compiled in Table S1. Image analysis All plaque images were independently analyzed by two certified neuroradiologists from the IsCore Image Corelab (http://imagecorelabcn.com/en/) using CMRtools software (Cardiovascular Imaging Solutions Ltd., London, UK). The original imaging data were required in Digital Imaging and Communications in Medicine (DICOM) format. Prior to formal assessment of the imaging data, a training session was conducted using 5% of the cohort data to familiarize the raters with the assessment criteria. Formal evaluation was initiated only after the inter-rater agreement reached an "excellent" level, defined as a reliability coefficient greater than 0.75. The evaluation specifically involved manual tracing of the outer vessel wall boundary (for calculating vessel area, VA) and the inner luminal boundary (for calculating lumen area, LA) on cross-sectional T1-weighted images of the BA at the maximal luminal narrowing (MLN) and reference levels after magnification. Measurements of luminal diameter, enhancement ratio, VA, and LA were performed at these specific locations. To assess plaque enhancement, the signal intensity ratio (SIR) was calculated as the signal intensity of the plaque (SIplaque) divided by that of the reference tissue (brain white matter) (SIreference). An SIR threshold of ≥1.5 was defined as plaque enhancement, whereas an SIR <1.5 was defined as non-plaque enhancement (NPE) (18). The MLN was defined as the single cross-section showing the most severe luminal stenosis, as confirmed jointly by DSA and HR-MRI. Reference sites were determined according to the criteria established in the Warfarin versus Aspirin for Symptomatic Intracranial Disease (WASID) trial: the normal vessel segment immediately proximal to the stenosis was preferentially selected; if the proximal segment was also diseased, the distal normal segment was used instead (19). The cross-sectional culprit lesion was divided into four quadrants: ventral, dorsal, left, and right (20). For analytical convenience, the left and right quadrants were combined into a single lateral category. In cases where a plaque extended across two quadrants with the thickest region spanning both, it was classified as involving more than two quadrants. Detailed definitions and measurement criteria for plaque characteristics are summarized in Table S2. Outcome assessment NIBL were assessed based on the presence of high signals on DWI and low signals on apparent diffusion coefficient (ADC) within 48 hours post-procedurally, compared with preoperative imaging. Patients demonstrating acute ischemic lesions within the BA territory, encompassing posterior fossa and supratentorial structures, were designated as DWI+ based on restricted diffusion criteria. Statistical analysis Analytic procedures were performed using the software SPSS 25.0 (IBM Corp., Armonk, NY, USA), with normally distributed variables summarized as mean ± standard deviation (SD), and non-normally distributed variables as medians with interquartile ranges or as counts with percentages. Descriptive analysis was conducted for all patients, categorizing them into DWI+ and DWI− groups. Continuous variables were compared using the t‑test or Mann‑Whitney U test, and categorical variables were compared using the χ2 test. For the multivariate logistic regression model, variables with a P value <0.10 in univariate analysis were entered as candidates, whereas age and sex were forcibly included regardless of their univariate P values due to their clinical importance. Multicollinearity among independent variables was assessed using the variance inflation factor (VIF), with a VIF >5 indicating significant collinearity. Stratified analysis by EVT modality was performed, and statistically significant factors were further analyzed using multivariate regression analysis to investigate their impact on NIBL across different EVT modalities. Statistical significance was defined using a two-tailed alpha level of 0.05. Results Preoperative baseline characteristics Figure 1 illustrates the study flowchart. A total of 373 patients met the inclusion criteria, among whom 188 developed NIBL, including 30 cases with symptomatic ischemic lesions. The mean age among participants was 61.76±7.83 years. Diabetes was found to be more prevalent in patients with NIBL compared with those without (51.6% vs. 40%, P=0.025). In contrast, patients without NIBL were more frequently associated with smoking (55.7% vs. 42%, P=0.008) and alcohol consumption (40% vs. 29.8%, P=0.038). Patients with a preoperative stroke were more likely to develop NIBL post-procedurally (70.7% vs. 61.1%, P=0.049) (Table 1). Flowchart of patient inclusion in the present study. BAS, basilar artery stenosis; DICOM, Digital Imaging and Communications in Medicine; EVT, endovascular therapy; HR-MRI, high-resolution magnetic resonance imaging; NIBL, new ischemic brain lesion. Table 1Baseline characteristicsCharacteristicAll participants (n=373)DWI+ group (n=188)DWI− group (n=185)P valueAge (years)61.8±7.861.7±7.461.9±8.20.821Men284 (76.1)142 (75.5)142 (76.8)0.781BMI (kg/m2)26.4±3.126.6±3.126.3±3.10.469Hypertension318 (85.3)166 (88.3)152 (82.2)0.095Diabetes mellitus171 (45.8)97 (51.6)74 (40.0)0.025Hyperlipidemia139 (37.3)71 (37.8)68 (36.8)0.840Coronary artery disease55 (14.7)28 (14.9)27 (14.6)0.935Atrial fibrillation6 (1.6)2 (1.1)4 (2.2)0.446Smoking history182 (48.8)79 (42.0)103 (55.7)0.008Drinking history130 (34.9)56 (29.8)74 (40.0)0.038Preoperative mRS ≤1373 (100.0)188 (100.0)185 (100.0)>0.99Qualifying events0.049 TIA127 (34.0)55 (29.3)72 (38.9) Ischemic stroke246 (66.0)133 (70.7)113 (61.1)LDL (mmol/L)1.9±0.61.9±0.71.8±0.60.628Values are presented as n (%), mean ± standard deviation. BMI, body mass index; DWI, diffusion-weighted imaging; LDL, low-density lipoprotein; mRS, modified Rankin Scale; TIA, transient ischemic attack. Baseline characteristics Values are presented as n (%), mean ± standard deviation. BMI, body mass index; DWI, diffusion-weighted imaging; LDL, low-density lipoprotein; mRS, modified Rankin Scale; TIA, transient ischemic attack. HR-MRI and angiography characteristics Significant differences were observed between the two groups in PR (51.1% vs. 37.8%, P=0.01), PE (87.2% vs. 75.7%, P=0.004), and plaque burden (19% vs. 12%, P=0.033). Conversely, the groups showed no statistically significant variations in stenosis severity or plaque distribution. Moreover, the incidence of NIBL post-procedurally did not vary significantly across treatment modalities (PA: 28.2% vs. 33.5%; BMS: 36.7% vs. 32.4%; SES: 35.1% vs. 34.1%) (Table 2). Figure 2 and Figure 3 present the preoperative HR-MRI, intraoperative DSA, and postoperative DWI images of two patients. Table 2Lesion characteristics of patients with BAS based on HR-MRI and DSACharacteristicAll participants (n=373)DWI+ group (n=188)DWI− group (n=185)P valueRemodeling type0.010 PR166 (44.5)96 (51.1)70 (37.8) NPR207 (55.5)92 (48.9)115 (62.2)PE304 (81.5)164 (87.2)140 (75.7)0.004Plaque burden (%)†16.4 (30.5)19 (34.9)12 (37.9)0.033Plaque eccentric263 (70.5)131 (69.7)132 (71.4)0.724Plaque distribution (axial view)0.881 Ventral137 (36.7)68 (36.2)69 (37.3) Dorsal58 (15.5)28 (14.9)30 (16.2) Lateral178 (47.7)92 (48.9)86 (46.5)Mori type0.192 A182 (48.8)93 (49.5)89 (48.1) B160 (42.9)75 (39.9)85 (45.9) C31 (8.3)20 (10.6)11 (5.9)Plaque distribution (coronal view)0.089 Low177 (47.5)81 (43.1)96 (51.9) Middle196 (52.5)107 (56.9)89 (48.1)Plaque length (mm)7.1±3.77.3±3.76.9±3.80.277Stenosis (%)78.5±8.578.6±8.078.5±9.10.987Treatment type0.502 PA115 (30.8)53 (28.2)62 (33.5) BMS129 (34.6)69 (36.7)60 (32.4) SES129 (34.6)66 (35.1)63 (34.1)Values are presented as n (%), mean ± standard deviation or median (interquartile range)†. BAS, basilar artery stenosis; BMS, balloon-mounted stents; DSA, digital subtraction angiography; DWI, diffusion-weighted imaging; HR-MRI, high-resolution magnetic resonance imaging; NPR, non-positive remodeling; PA, primary angioplasty; PE, plaque enhancement; PR, positive remodeling; SES, self-expansion stent. Lesion characteristics of patients with BAS based on HR-MRI and DSA Values are presented as n (%), mean ± standard deviation or median (interquartile range)†. BAS, basilar artery stenosis; BMS, balloon-mounted stents; DSA, digital subtraction angiography; DWI, diffusion-weighted imaging; HR-MRI, high-resolution magnetic resonance imaging; NPR, non-positive remodeling; PA, primary angioplasty; PE, plaque enhancement; PR, positive remodeling; SES, self-expansion stent. A 76-year-old male patient with a smoking history presented with a 3-week history of dizziness. (A) Preoperative DSA showed a 90% stenosis in the middle segment of the basilar artery. (B) The stenotic segment was treated with a single inflation using a 2.25×25 mm balloon (NEURO RX, SINOMED). Within 72 hours after the operation, the patient developed a new asymptomatic cerebral infarction in the right cerebellum. DWI revealed a new ischemic brain lesion (C) and ADC imaging showed a new low-signal lesion (F). (D) and (E) are cross-sectional contrast-enhanced T1-weighted images at the site of MLN and the REF site of the basilar artery, respectively. A non-eccentric plaque can be observed in (D). After magnification of the images, the VA and LA were manually measured at the MLN site (D: VA 17.87 mm2, LA 1.07 mm2) and the REF site (E: VA 16.64 mm2, LA 3.80 mm2). In this case, the plaque burden was calculated to be 22.16%, and the remodeling index at the MLN site was 1.07, indicating positive remodeling. ADC, apparent diffusion coefficient; DSA, digital subtraction angiography; DWI, diffusion-weighted imaging; LA, lumen area; MLN, maximal lumen narrowing; REF, reference; VA, vessel area. A 62-year-old male patient with a history of smoking and alcohol consumption presented with recurrent episodic dizziness for 1 month. (A) Preoperative DSA revealed stenosis in the middle segment of the basilar artery, with a degree of narrowing of 80%. (B) The stenotic segment was treated with one inflation using a 1.5×15 mm balloon (NEURO RX, SINOMED), followed by another inflation with a 2.5×20 mm balloon (Maverick, Boston Scientific). Within 72 hours after the procedure, the patient developed a new asymptomatic cerebral infarction in the left cerebellum. DWI revealed a new ischemic brain lesion (C), and ADC imaging showed a new low-signal lesion (F). (D) and (E) are cross-sectional contrast-enhanced T1-weighted images at the site of MLN and the REF of the basilar artery, respectively. Plaque enhancement is visible, showing an eccentric morphology and located on the dorsal side of the basilar artery (D). After magnifying the images, the VA and LA were manually traced and measured at the MLN site (D: VA 24.69 mm2, LA 1.77 mm2) and the REF site (E: VA 26.70 mm2, LA 11.32 mm2). In this case, the plaque burden was 30.54%, and the remodeling index at the MLN site was 0.92, indicating non-positive remodeling. ADC, apparent diffusion coefficient; DSA, digital subtraction angiography; DWI, diffusion-weighted imaging; LA, lumen area; MLN, maximal lumen narrowing; REF, reference; VA, vessel area. Risk factor analysis for NIBL Based on conventional statistical results, covariates such as remodeling type, PE, plaque distribution, eccentricity, treatment modality, degree of stenosis, and Mori classification were entered into the multivariable regression model. Multicollinearity among these variables was assessed using the VIF, and the results showed that plaque burden, remodeling index, and degree of stenosis had VIF values of 2.26, 2.25, and 1.00, respectively, all below the threshold of 5, indicating no significant collinearity. According to multivariate logistic regression analysis, both PE [odds ratio (OR) =2.09, 95% confidence interval (CI): 1.18–3.70, P=0.011] and PR (OR =1.77, 95% CI: 1.03–3.06, P=0.039) emerged as independent predictors of NIBL (Table 3). Table 3Multivariable logistic regression of predictors of new ischemic brain lesions after EVT for patients with BASCharacteristicUnadjusted PAdjusted† PAdjusted OR (95% CI)Plaque characteristic PR0.0100.0391.77 (1.03, 3.06) PE0.0050.0112.09 (1.18, 3.70)Treatment type PA––Ref BMS0.2490.1011.58 (0.92, 2.73) SES0.4290.4101.25 (0.73, 2.13)†, outcomes adjusted by age, smoking, drinking, hypertension, diabetes mellitus, qualifying events, plaque burden (%), and plaque distribution (coronal view). BAS, basilar artery stenosis; BMS, balloon-mounted stents; CI, confidence interval; EVT, endovascular therapy; OR, odds ratio; PA, primary angioplasty; PE, plaque enhancement; PR, positive remodeling; Ref, reference; SES, self-expansion stent. Multivariable logistic regression of predictors of new ischemic brain lesions after EVT for patients with BAS †, outcomes adjusted by age, smoking, drinking, hypertension, diabetes mellitus, qualifying events, plaque burden (%), and plaque distribution (coronal view). BAS, basilar artery stenosis; BMS, balloon-mounted stents; CI, confidence interval; EVT, endovascular therapy; OR, odds ratio; PA, primary angioplasty; PE, plaque enhancement; PR, positive remodeling; Ref, reference; SES, self-expansion stent. Subgroup analysis To guide more precise treatment modalities based on patients’ risk factors, we categorized patients according to PE and vessel remodeling types (Figure 4). We observed that in the non-enhanced subgroup, BMS treatment was associated with a higher incidence of NIBL compared with PA and SES [BMS: 50% (11/22) vs. PA: 26.1% (6/23) vs. SES: 29.2% (7/24)]. Compared with patients without risk factors, those with both PE and PR had a higher incidence of NIBL post-procedurally following PA and SES [PA: 63.2% (24/38) vs. 23.1% (3/13); SES: 67.4% (31/46) vs. 28.6% (4/14)]. However, the mechanisms of NIBL did not differ significantly across treatment modalities. We also found that the incidence of NIBL was lowest in patients without PE or PR, regardless of treatment modality. The specific data are presented in Tables S3,S4. Incidence of NIBL in patients with different plaque characteristic combinations under various treatment modalities. BMS, balloon-mounted stents; EVT, endovascular therapy; NIBL, new ischemic brain lesion; NPE, non-plaque enhancement; NPR, non-positive remodeling; PA, primary angioplasty; PE, plaque enhancement; PR, positive remodeling; SES, self-expanding stents. Discussion This study employed HR-MRI and represents one of the largest investigations to date in terms of sample size. It systematically analyzed risk factors for NIBL following EVT for BAS, while highlighting the significant association between HR-MRI-identified vulnerable plaque markers and NIBL. By focusing on PE and PR as key indicators of plaque instability, we elucidated their role in predicting acute ischemic lesions captured by DWI, as well as differences in the incidence of NIBL across different treatment modalities. Compared with postprocedural symptomatic infarctions, NIBL more accurately reflect clinical outcomes after EVT. Given that postprocedural NIBL is a common complication following endovascular procedures, most EVT studies (such as the SAMMPRIS and CASSISS trials) have established postprocedural new cerebral ischemic events as study endpoints, yet primarily focused on symptomatic ischemic events (6,21). Some new ischemic lesions are asymptomatic yet carry a risk of increased future stroke (2). Studies have reported that the incidence of new DWI abnormalities after intracranial EVT can be as high as 65%, but these studies did not specifically analyze the BA (22). The present study demonstrated that the incidence of NIBL after EVT for BAS was 50.4%. Our study demonstrated that PE and PR among the HR-MRI features of vulnerable plaques are independent risk factors for NIBL following endovascular therapy for BAS. Prior research has indicated that PE serves as a predictor for subsequent cerebral ischemic events (23). Enhanced signals within plaques are also considered an effective marker for distinguishing culprit from non-culprit plaques (24). Research indicates that enhanced plaques contain significantly more macrophages and exhibit greater inflammation compared with non-enhanced plaques (25). The inflammation throughout the plaque’s life cycle promotes the thinning of the fibrous cap, and may even lead to ulceration and erosion on the plaque surface (26). During EVT, these vulnerable plaques are more prone to rupture due to external mechanical forces, resulting in thrombosis and NIBL post-procedurally. Meanwhile, PR helps to maintain lumen size in atherosclerosis (27). Compared to anterior circulation vessels, posterior circulation vessels more frequently exhibit PR (28). PR reduces local shear stress, promotes collagen and elastin degradation, and leads to arterial dilation, increasing plaque burden (29). Our previous work found that the remodeling index is closely linked to plaque burden, with an inverse relationship between them. A higher plaque burden is associated with a greater potential risk of fibrous cap rupture (16). Therefore, BAS patients with PE and PR are more likely to have plaque rupture during surgery and lead to NIBL. Previous studies have indicated no significant differences in symptomatic stroke, transient ischemic attack (TIA), or one-year mortality among different EVT approaches for intracranial BAS (30). However, our study found that the incidence of NIBL varied among patients receiving different EVT modalities depending on specific plaque characteristics, although this difference did not reach statistical significance. In both PA and SES groups, patients with both PE and PR had the highest incidence of postprocedural NIBL, whereas those without PE and PR had the lowest NIBL incidence [PA: 63.2% (24/38) vs. 23.1% (3/13); SES: 67.4% (31/46) vs. 28.6% (4/14)]. In the BMS group, patients with only PR had the highest NIBL incidence, and those without PE and PR also had the lowest NIBL incidence, but the difference was not significant [57.1% (4/7) vs. 46.7% (7/15)]. The direct compression of plaques by simple PA and the cutting effect of SES on large-volume unstable plaques are the main contributors to the occurrence of NIBL. BMS may enhance the cutting effect on plaques due to their deployment method, leading to a similar probability of NIBL in patients without PE and PR as in those with risk factors. In the future, clinicians may consider plaque characteristics when selecting treatment strategies to reduce the risk of postprocedural NIBL. This study has several limitations. Although it represents the largest investigation to date utilizing HR-MRI to examine risk factors for NIBL following EVT in BAS patients, the sample size remains relatively insufficient. Furthermore, this study exclusively enrolled symptomatic BAS patients who underwent EVT. Due to sample size limitations, no statistically significant differences in the incidence of NIBL were detected among different treatment approaches. We plan to validate these findings in larger-scale studies. Conclusions The above analysis demonstrates that PE and PR identified by HR-MRI in patients with BAS are independently associated with NIBL following EVT. These baseline plaque characteristics on HR-MRI not only hold predictive value for postprocedural ischemic events but also have the potential to serve as important imaging markers to assist in individualized selection of EVT strategies, thereby contributing to the reduction of postprocedural complications and improvement of clinical outcomes. Supplementary The article’s supplementary files as