Beyond Venous Inflammation: Vessel Wall Imaging of Intracranial Arteries in Neuro‐Behçet's Syndrome Authors: Esra Kochan Kizilkilic, Bora Korkmazer, Nursena Erener, Ahmet Kursat Karaman, Abdullah Arcan, Serdar Arslan, Aksel Siva, Emire Seyahi, Ugur Uygunoglu, Osman Kizilkilic Published: 2026 ABSTRACT ABSTRACTBackgroundNeuro‐Behçet syndrome (NBS) is a heterogeneous inflammatory disorder in which venous involvement is well recognized, while intracranial arterial pathology remains insufficiently characterized. Conventional luminal imaging methods are limited in detecting early or isolated arterial wall abnormalities. Vessel wall imaging (VWI) may provide additional insights into arterial involvement in NBS.MethodsThirty‐nine NBS patients (27 men, 12 women; mean age 38.52 ± 9.54 years) underwent intracranial VWI; 16 were imaged during the attack period and 23 during the attack‐free period. A total of 1170 artery segments were systematically evaluated for wall thickening patterns and contrast enhancement. VWI findings were analyzed in relation to imaging timing, radiological subtype, MR angiography findings, and EDSS.ResultsVessel wall involvement was found in 16 patients (41%), mainly in the posterior circulation (81%). Concentric wall thickening without enhancement was the most common pattern (80.48%). Vessel wall involvement was significantly more frequent in patients imaged during the attack period compared with those imaged in the attack‐free period (68.8% vs. 21.7%; p = 0.007). Vessel wall enhancement was present only in the attack period (p = 0.022). Notably, higher EDSS scores were significantly associated with multisegmental posterior circulation involvement, particularly when ≥ 3 segments were affected (p = 0.015; p = 0.024).ConclusionIntracranial arterial wall involvement is detectable by VWI in a substantial proportion of NBS patients, predominantly in the posterior circulation with a concentric thickening pattern. These findings suggest that arterial wall pathology may be more common than previously recognized and that VWI provides complementary diagnostic information beyond conventional luminal imaging. Introduction Behçet syndrome (BS), a chronic, multisystemic, inflammatory vasculitis, is characterized by recurrent oral aphthous ulcers, genital ulcers, and uveitis [1]. The mechanisms underlying its pathogenesis have not been elucidated. The disease is thought to involve both autoinflammatory and autoimmune mechanisms. BS is one of the rare vasculitides affecting venous and arterial systems with a distinct predilection for veins [2]. In patients with BS, inflammation can lead to vessel wall thickening, stenosis, aneurysm formation, and thrombus formation in both arteries and veins [3, 4]. Neuro‐Behçet syndrome (NBS), defined by the development of neurological manifestations in the setting of BS, occurs in approximately 5%–10% of patients [5]. The lesions in NBS are frequently located in the brainstem and diencephalon, resulting from inflammation of small venules. The prevalence rate of arterial involvement in patients with NBS is 1%–5%. Arterial involvement manifests mostly as small‐vessel vasculitis. However, case reports have also reported narrowing and occlusion of medium‐sized and large‐sized intracranial arteries [6, 7, 8, 9]. Vascular heterogeneity, which indicates that the vessels of varying sizes in both the venous and arterial systems are affected, is a major challenge in elucidating the pathophysiology of NBS. Current imaging methods focus on the vessel lumen and can only reveal narrowing or occlusion. However, vessel wall imaging (VWI) enables the examination beyond the vessel lumen, revealing isolated wall thickening, especially in early‐stage vasculitis. Recent studies suggest that VWI enables the differential diagnosis of intracranial vasculopathies, especially when combined with traditional luminal imaging methods [10]. Limited cases of intracranial arterial involvement due to BS have been reported using VWI [11, 12]. This is the first study to systematically investigate the application of VWI in NBS. Methods Patients and Clinical Evaluation In this exploratory pilot study, vessel wall imaging of patients diagnosed with NBS was evaluated. The inclusion criteria were as follows: patients aged 18–65 years; patients who met the International Study Group criteria for BS [13]; patients with clinical, imaging, and laboratory features consistent with NBS [14]; patients who provided informed consent. The exclusion criteria were as follows: patients with major vascular risk factors for intracranial atherosclerosis (diabetes mellitus, hypertension, metabolic syndrome, and hyperlipidemia); patients with known systemic vasculitis (giant cell arteritis, Takayasu arteritis, and polyarteritis nodosa); patients with other rheumatological diseases (systemic lupus erythematosus and Sjögren's syndrome); patients with active systemic infection; patients with a history of ischemic stroke; patients with contraindications to magnetic resonance imaging (MRI) or gadolinium‐based contrast agents. The data on demographic characteristics, age at diagnosis of BS and NBS, and disease duration were recorded. Neurological disability was quantified using the Expanded Disability Status Scale (EDSS), a tool used to quantify disability in multiple sclerosis [15]. All assessments were performed by experienced neurologists (E.K.K. and N.E.) trained in EDSS evaluation. Patients were divided into two subgroups based on imaging timing. The first group included patients who underwent VWI during the attack period, defined as imaging within 3 months of symptom onset. This threshold was chosen to capture the acute/subacute inflammatory phase, as vessel wall enhancement in vasculitis reflects active inflammation and tends to diminish over time or with treatment. The second group comprised patients consecutively included during routine outpatient follow‐up who underwent VWI in the attack‐free period. Radiological and Procedural Evaluations Vessel Wall MRI Protocol MRI examinations were performed using a 3 Tesla MR system (Ingenia; Philips Healthcare, Best, Netherlands) and a 32‐channel head coil. The standard MRI VWI protocol used in this study was pre‐contrast and post‐contrast sagittal 3D Brain VIEW with improved motion‐sensitized driven‐equilibrium (TR/TE, 700/35 ms; flip angle, 80°; field of view: 200 × 251 × 183 mm; acquisition spatial resolution: 0.8 × 0.8 × 0.8 mm3; reconstructed spatial resolution: 0.58 × 0.58 × 0.4 mm3; acquisition time: 5.11 min). The MRI protocol included time of flight (TOF) magnetic resonance venography, TOF magnetic resonance angiography (MRA), axial T2‐weighted (T2W), axial fluid‐attenuated inversion recovery, and diffusion‐weighted imaging sequences, along with VWI sequences. Post‐contrast 3D Brain VIEW sequences were captured after intravenously injecting gadobutrol (0.1 mmol/kg body weight; Gadovist; Bayer Schering Pharma, Berlin, Germany). Radiological Evaluation MRI patterns were independently reviewed by two neuroradiologists (S.A. and O.K.), who were blinded to clinical findings and had 15 and 25 years of experience in neuroimaging, respectively. Discrepancies were resolved by consensus after discussion, and the consensus assessment was used for the final analysis. To descriptively capture the heterogeneity of radiological presentations in NBS and to allow comparison of VWI findings across different imaging patterns, imaging findings were grouped into four radiological patterns: isolated cerebral venous sinus thrombosis (CVST); typical parenchymal inflammatory lesions; arterial lesions; and combined typical parenchymal inflammatory and arterial lesions. In this study, the arterial lesions subgroup was defined as atypical NBS, which includes parenchymal imaging findings inconsistent with the classic inflammatory NBS pattern. This group includes isolated great artery stenosis or occlusion, as well as cortical, subcortical, or atypical white matter lesions that do not show typical brainstem‐diencephalic involvement. Inflammatory parenchymal lesions were classified according to their anatomical location as telencephalic, diencephalic, brainstem and/or cerebellar, and spinal cord. Lesions involving more than one of these regions were defined as combined forms. For the evaluation of arterial involvement, intracranial arteries were divided into 30 separate segments (V4 segments of the vertebral artery, basilar artery, and basilar artery bifurcation; P1, P2, and P3 segments of the posterior cerebral artery (PCA), posterior communicating artery, supraclinoid segment of the internal carotid artery (ICA), and communicating segment of the ICA; M1, M2, and M3 segments of the middle cerebral artery (MCA), MCA bifurcation, and anterior communicating artery; A1, A2, and A3 segments of the anterior cerebral artery (ACA)). The percentage of stenosis for each segment on MRA was calculated using the WASID method (stenosis (%) = (1−(maximal stenotic artery diameter/Proximal healthy artery diameter)) × 100) [16]. Each artery segment was classified into five categories according to the stenosis percentage: Healthy: 0%–9%; mild stenosis: 10%–29%; moderate stenosis: 30%–69%; severe stenosis: 70%–99%; occlusion: 100%. The presence of intracranial aneurysms was recorded. Arterial lesions were recorded as present or absent without further etiological or morphological subcategorization. The vessel wall of the intracranial artery segments was evaluated using multiplanar reconstruction on the 3D pre‐contrast and post‐contrast T1W VWI sequences. Each artery segment was grouped into the following five groups according to the vessel wall thickening pattern and the presence or absence of wall enhancement: Group 1: Physiological appearance (no wall thickening or enhancement) Group 2: Concentric wall thickening with wall enhancement Group 2A: Only concentric wall thickening (without enhancement) Group 3: Eccentric wall thickening with wall enhancement Group 3A: Only eccentric wall thickening (without enhancement) The lesions were defined as follows: concentric lesions: the thinnest wall segment width ≥ 50% of the thickest segment on VWI; eccentric lesions: the thinnest segment width < 50% of the thickest segment on VWI. Pre‐contrast VWI and post‐contrast VWI were compared to evaluate wall enhancement. Statistical Analysis All statistical analyses were performed using SPSS (version 28.0 software) (IBM, Armonk, NY, USA). Categorical data are presented as percentages (%), while continuous data are presented as mean ± SD. Analyses were performed using χ 2‐test, Student's t‐test, ANOVA, and Pearson correlation tests following the data properties. Binary logistic regression was performed to evaluate the independent effects of gender and attack status at the time of imaging (variables found to be associated in univariate analyses with p < 0.05) on the presence of vessel wall involvement. Because the sample size was larger than 30, the normal distribution assumption was neglected in the parametric tests based on the central limit theorem [17]. Statistical significance was set at p < 0.05, and absolute p‐values were given for each analysis. No formal correction for multiple comparisons was applied due to the exploratory design of the study. Ethical Approval The study was approved by the Clinical Research Ethics Committee of Istanbul University‐Cerrahpaşa (Ethics ID: 960814, 2024). All volunteers were provided with information about the procedures. Written informed consent was obtained from all participants. The study was performed according to the principles of the Declaration of Helsinki. Results Demographic and Clinical Characteristics This study enrolled 39 patients diagnosed with NBS (27 males (69.2%) and 12 females (30.8%)). The demographic and clinical characteristics of the patients are summarized in Table 1. The mean age at study entry was 38.52 ± 9.54 years, with a mean age at BS and NBS diagnosis of 30.61 ± 8.20 and 35 ± 10.35 years, respectively. The mean EDSS score at the time of the study was 2.93 ± 2.22. Apart from two patients who were off treatment, all were receiving immunosuppressive therapy, most commonly anti–tumor necrosis factor agents, alone or in combination with azathioprine, low‐dose prednisolone, and/or colchicine. Sixteen patients (41.0%; 8 males and 8 females) underwent imaging during an attack period, whereas 23 were evaluated during an attack‐free period. The interval between symptom onset and imaging was calculated as 31.3 ± 24.4 days (range: 3–90). Female patients were significantly overrepresented in the attack group compared with the attack‐free group (8/16 vs. 4/23, p = 0.041). Other demographic and clinical variables—including age at study entry, age at diagnosis of BS or NBS, and EDSS score—did not differ significantly between those presented with and without attack (Table S1). TABLE 1Comparison of demographic and clinical features according to vessel wall involvement.VariableTotal (n = 39)Vessel wall involvement (+) (n = 16)Vessel wall involvement (−) (n = 23) p Current age, mean ± SD, years38.52 ± 9.5441.68 ± 7.7136.21 ± 10.380.081Male sex, n (%)27 (69.2%)8 (50%)19 (82.6%)0.041* Age at diagnosis of BS, mean ± SD, years30.61 ± 8.2032.25 ± 6.5729.48 ± 9.360.307Duration of BS, mean ± SD, years8.18 ± 7.669.43 ± 7.927.30 ± 7.530.405Age at diagnosis of NBS, mean ± SD, years35 ± 10.3537.18 ± 9.8233.26 ± 10.760.271Duration of NBS, mean ± SD, years3.48 ± 5.104.43 ± 5.912.87 ± 2.620.262EDSS, mean ± SD2.93 ± 2.223.03 ± 1.562.87 ± 2.620.827Imaging during attack period, n (%)16 (41.0)11 (68.8)5 (21.7)0.007* Immunosuppressive therapy, n (%)37 (94.9)16/1621/230.503 Note: Data are presented as mean ± standard deviation or n (%).Abbreviations: BS, Behçet syndrome; EDSS, Expanded Disability Status Scale; NBS, Neuro‐Behçet syndrome; SD, standard deviation.* p < 0.05. Comparison of demographic and clinical features according to vessel wall involvement. Note: Data are presented as mean ± standard deviation or n (%). Abbreviations: BS, Behçet syndrome; EDSS, Expanded Disability Status Scale; NBS, Neuro‐Behçet syndrome; SD, standard deviation. p < 0.05. Radiological NBS Subtypes Patients were categorized into four radiological NBS subtypes according to predominant imaging findings. Of the 39 patients, 3 (7.7%) had isolated CVST, 28 (71.8%) had typical parenchymal inflammatory lesions, 4 (10.3%) had arterial lesions, and 4 (10.3%) had combined lesions (typical parenchymal inflammatory and arterial). Distribution of Parenchymal Lesions Analysis of parenchymal inflammatory lesion distribution showed that 7 patients (17.9%) had no parenchymal inflammatory lesions, including 4 with arterial lesions and 3 with isolated CVST subtypes. Among the remaining 32 patients, at least one neuroanatomical region was involved. Brainstem and/or cerebellar involvement was most frequent (n = 27, 69.2%), followed by diencephalic involvement (n = 13, 33.3%) and spinal cord involvement (n = 3, 7.7%). Within the diencephalic subgroup (n = 13), 6 patients (15.4%) had combined diencephalic plus brainstem/cerebellar involvement, 5 (12.8%) showed extensive involvement of the telencephalon, diencephalon, and brainstem/cerebellum, and 2 (5.1%) had isolated diencephalic involvement. Vessel Wall Involvement As shown in Table 1, vessel wall involvement was identified in 16 of 39 patients (41.0%), whereas vessel wall enhancement was present in only 4 patients (10.3%). Anatomic Distribution Among patients with vessel wall involvement, posterior circulation involvement predominated (15/16, 93.75%), while anterior circulation involvement was observed in 6/16 patients (37.5%); combined anterior and posterior involvement was present in 5/16 patients (31.25%). Across all patients, a total of 1170 vascular segments (30 segments per patient) were evaluated for vessel wall involvement. Involvement was detected in 41 segments overall (3.5%), of which 33 (80.48%) were in the posterior circulation and 8 (19.5%) in the anterior circulation. The most frequently affected site was the P1 segment of the PCA with 14 segments (34.1%) exhibiting involvement. According to the pattern of vessel wall involvement, the most affected segments were observed in group 2A (33 segments; 80.48%), followed by group 2 (5 segments, 12.2%), group 3 (1 segment, 2.4%), and group 3A (2 segments, 4.9%). Table 2 shows the distribution and involvement pattern of vessel wall lesions in intracranial artery segments. Representative cases are illustrated in Figures 1 and 2. TABLE 2Locations of vessel wall lesions and observed wall involvement patterns on VWI. Vessel wall lesion location (total involved segments, n = 41) PCA P114 (34.1%)Basilar artery bifurcation6 (14.7%)Vertebral V46 (14.7%)MCA M13 (7.3%)Basilar artery3 (7.3%)PCA P23 (7.3%)ICA supraclinoid2 (4.9%)ICA communicating2 (4.9%)MCA M21 (2.4%)PCA P31 (2.4%) Vessel wall involvement patterns Group 2A33 (80.48%)Group 25 (12.19%)Group 3A2 (4.87%)Group 31 (2.43%) Note: Group 2A: concentric wall thickening, Group 2: concentric wall thickening with wall enhancement, Group 3A: eccentric wall thickening, Group 3: eccentric wall thickening with wall enhancement.Abbreviations: ACA, anterior cerebral artery; ICA, internal carotid artery; MCA, middle cerebral artery; PCA, posterior cerebral artery; VWI, Vessel wall imaging. Locations of vessel wall lesions and observed wall involvement patterns on VWI. Note: Group 2A: concentric wall thickening, Group 2: concentric wall thickening with wall enhancement, Group 3A: eccentric wall thickening, Group 3: eccentric wall thickening with wall enhancement. Abbreviations: ACA, anterior cerebral artery; ICA, internal carotid artery; MCA, middle cerebral artery; PCA, posterior cerebral artery; VWI, Vessel wall imaging. Active inflammatory brainstem lesion associated with vessel wall involvement. (a) Axial T2 weighted and (b) Axial FLAIR images showing right pontomesencephalic junction and left superior cerebellar peduncle hyperintense lesions. (c) Axial precontrast and (d) Axial postcontrast images showing contrast enhancement of both active inflammatory lesions. (e) Axial precontrast vessel wall image and (f). Axial postcontrast vessel wall images at mesencephalic level showing wall thickening (arrow) at bilateral PCA P1 segment without contrast enhancement and bilateral mesencephalic parenchymal enhancements of the inflammatory lesions as well. Basal ganglia chronic ischemic lesion associated with vessel wall enhancement of posterior circulation. (a) Axial TSE T2 weighted image, (b) Axial FLAIR image, (c) Axial SWI image and (d) Coronal TSE T2 weighted images showing hemorrhagic ischemic lesion in the right globus pallidus and chronic ischemic lesion in the left globus pallidus. (e) Axial pre‐contrast black blood vessel wall imaging and (f) Axial postcontrast black blood vessel wall imaging demonstrating wall thickening and wall enhancement of the left PCA P2 segment. Comparative Analysis of Patients With and Without Vessel Wall Involvement Vessel wall involvement was more frequently detected in female patients compared to male patients (8/12 [66.7%] vs. 8/27 [29.6%]; p = 0.041). Additionally, vessel wall involvement was significantly associated with imaging performed during the attack period; 11 of 16 patients (68.8%) with vessel wall involvement were imaged during an attack, compared to only 5 of 23 (21.7%) without involvement (p = 0.007) (Table 1). No significant differences were observed between groups regarding current age, age at diagnosis of BS or NBS, disease duration, EDSS scores, or immunosuppressive therapy use. Binary logistic regression was performed to evaluate the independent effects of gender and attack status at the time of imaging on the presence of vessel wall involvement. Attack status emerged as a significant independent predictor of vessel wall involvement (OR = 6.14, 95% CI: 1.36–27.63, p = 0.018). Female gender showed a trend (OR = 3.01, 95% CI: 0.60–15.01, p = 0.180), though this did not reach statistical significance. When analyzed according to radiological subtypes, vessel wall involvement was absent in all patients with isolated CVST (0/3), while it was present in 10 of 28 patients (35.7%) with typical parenchymal inflammatory lesions, 3 of 4 with arterial lesions, and 3 of 4 with combined parenchymal inflammatory and arterial lesions. Notably, three cases of pure spinal NBS were identified within the parenchymal inflammatory subtype, one of which demonstrated multisegmental vessel wall involvement. Evaluation of Anterior/Posterior System Involvement in Vessel Wall Involvement Among the 16 patients with vessel wall involvement, no significant associations were observed between anterior or posterior vessel wall involvement and radiological subtype or parenchymal lesion location. However, in an exploratory analysis, the number of posterior circulation segments involved showed a positive correlation with EDSS scores (r = 0.594, p = 0.015). Patients with involvement ≥ 3 posterior segments had higher EDSS scores than those with < 3 segments (2.86 ± 1.34 vs. 1.44 ± 0.88; p = 0.024). Vessel Wall Enhancement Vessel wall enhancement was observed in 4 of 39 patients (10.3%), all of whom were imaged during the attack period. Among the 16 patients imaged during an attack, 4 (25.0%) demonstrated vessel wall enhancement, whereas none of the 23 patients imaged during the attack‐free period showed enhancement (p = 0.022). Correlation of MRA Findings With VWI Findings MRA abnormalities were detected in 5 patients (12.8%) (2 aneurysm cases and 3 arterial stenosis cases). The involvement of the same segment of the vessel wall was detected in 2 of 5 patients with MRA abnormalities. The data are summarized in Table S2. The flowchart in Figure 3 summarizes the radiological subgroups of patients and their vessel wall involvement patterns in attack versus attack‐free periods. Flow chart summarizing the radiological subgroups of patients and their vessel wall involvement patterns in attack versus attack‐free periods. Discussion This study shows that intracranial arterial wall involvement is detectable in 41% of patients with NBS when VWI is systematically applied. The predominant pattern was concentric wall thickening without enhancement, with a clear predilection for the posterior circulation—most commonly the P1 segment of the PCA. Vessel wall involvement was significantly more frequent during the attack period. Wall enhancement was present in a few patients and was exclusively observed during attack periods. Furthermore, multisegmental posterior circulation involvement was associated with higher disability scores. These findings suggest that VWI may have potential value as an adjunctive imaging tool in NBS. NBS manifests in two well‐characterized forms: the classic parenchymal form (p‐NBS), accounting for 75%–80% of cases, and CVST, seen in 20%–25% [14]. The pathophysiology of both forms is rooted in venous involvement. In p‐NBS, the characteristic parenchymal lesions, typically located at the meso‐diencephalic junction and extending along brainstem tracts, correspond to the anatomical distribution of intra‐axial veins rather than arterial territories [18]. Perivascular lymphocytic infiltration centered on small venules is the hallmark pathological finding [19, 20]. Similarly, CVST reflects involvement of the cerebral venous sinuses—structures that are functionally analogous to veins. This venous predilection is consistent with the broader vascular profile of BS, in which deep venous thrombosis is the most frequent systemic vascular manifestation, occurring in 60%–80% of cases [5, 19, 20]. Despite this venous predominance, central nervous system arterial involvement is rare and has been documented mainly through case reports and small series. The reported manifestations include intracranial aneurysms [21, 22] and arterial stenosis or occlusion [8, 9, 23, 24], frequently affecting the proximal segments of medium‐sized to large‐sized vessels, such as the MCA and basilar artery. Only two case studies have reported VWI findings in BS‐associated intracranial arteritis, both describing concentric wall thickening with contrast enhancement in anterior circulation vessels [11, 12]. Against this limited background, our findings suggest that arterial pathology in NBS may be substantially more common than previously appreciated, and that the diagnostic yield of conventional MRA may be significantly enhanced when combined with VWI. A central finding of the present study is the marked posterior circulation predominance of vessel wall involvement. The P1 segment of the PCA was the single most commonly involved site, followed by the basilar artery bifurcation and vertebral V4 segment. This posterior predilection is noteworthy and aligns with the well‐established clinical and radiological observation that parenchymal NBS lesions preferentially involve the brainstem and diencephalon, structures supplied by the vertebrobasilar circulation. The co‐localization of both parenchymal inflammatory lesions and intracranial arterial wall pathology within the posterior circulation territory suggests a shared pathophysiological milieu. These regions are supplied by short perforating arteries that are anatomically and functionally closely associated with the perivenular network, which represents the primary target of inflammation in NBS. Such an anatomical configuration may allow perivenular inflammatory processes to secondarily affect adjacent arterial walls without producing overt luminal abnormalities, resulting in predominantly concentric, non‐enhancing vessel wall thickening on VWI. In contrast, a primary arterial vasculitis would be expected to preferentially involve larger anterior circulation vessels, such as the MCA or ACA, which were not observed in our cohort. The distribution of NBS subtypes in this study was consistent with previous reports [18, 19, 25], with inflammatory parenchymal involvement being the most common, followed by relatively rare isolated CVST and large arterial subtypes. Vessel wall involvement was somewhat more frequent in the arterial than in the typical parenchymal inflammatory subtype and was absent in isolated CVST cases—suggesting a possible trend toward more pronounced vessel wall pathology in subtypes with an arterial component. Notably, among three pure spinal NBS cases, one demonstrated multisegmental intracranial posterior circulation vessel wall involvement, suggesting that cranial arterial pathology may extend beyond the clinical presentation even in isolated spinal disease. Anterior–posterior vessel wall involvement was not correlated with radiological subtypes. However, patients with multisegmental posterior involvement exhibited increased EDSS scores, suggesting an increased risk of clinical disability, especially when ≥ 3 segments are involved. However, due to the small and imbalanced NBS subtypes and the multiple subgroup comparisons, these findings should be interpreted with caution and considered exploratory. Other possible contributing factors, such as treatment‐related effects or nonspecific vessel wall changes, should also be considered in the interpretation of the observed findings. The predominant VWI pattern observed in our cohort was concentric thickening, with eccentric lesions accounting for only a small minority of affected segments. This is consistent with the established VWI literature on CNS vasculitis [26]. This observation, coupled with the young age of the cohort and the exclusion of patients with conventional atherosclerotic risk factors, may be more consistent with an inflammatory rather than atherosclerotic or dissection‐related mechanism underlying the observed vessel wall changes in NBS. In our study, VWI was significantly more frequent in patients imaged during the attack period, and wall enhancement was specifically detected in the attack period, albeit with low frequency. These findings indicate that intracranial arterial wall pathology in NBS is dynamic and temporally linked to disease activity: wall enhancement appears during the attack period and likely resolves during remission, while structural concentric thickening may persist. Vessel wall contrast enhancement is widely regarded as a hallmark of active vasculitis [26, 27, 28, 29]. This pattern has parallels in other vasculitides: in large‐vessel vasculitis, mural enhancement on MRI correlates with active inflammation and diminishes with treatment, though structural wall thickening may persist even in clinical remission [30, 31]. This interpretation is consistent with previous observations that vessel wall enhancement in CNS vasculitis can regress on follow‐up imaging while structural wall changes persist [27, 28]. This dynamic behavior raises the possibility that VWI may serve as a noninvasive biomarker of disease activity in NBS, complementing conventional MRI and clinical assessments. The low frequency of contrast enhancement in the study cohort differs from that reported in broader CNS vasculitis literature. Several factors likely contribute to this observation. The ongoing immunosuppressive therapy received by nearly all patients may have attenuated the inflammatory component, reducing the likelihood of detecting enhancement. Additionally, given the low incidence of NBS, our cohort was inherently heterogeneous with respect to disease phase, and the number of patients imaged during the attack period was limited. In addition, the relatively broad definition of the attack period (within 3 months) may have further contributed to this finding, as imaging performed later in this window could have missed peak inflammatory activity. This may, at least in part, explain the relatively low frequency of vessel wall enhancement observed in our cohort. The 3‐month window used to define the attack period requires further discussion. While many vasculitis studies perform acute imaging within days to weeks of symptom onset, this broader threshold was selected to capture both acute and subacute phases of inflammatory activity in NBS, where access to emergent high‐field VWI is not always feasible. This approach is supported by prior evidence that vessel wall enhancement in CNS vasculitis may persist well beyond the acute phase, with studies reporting stable or continued enhancement at follow‐up periods ranging from 8 to 13.5 months before eventual regression [27, 28, 32]. Nevertheless, a 3‐month window may have included patients in whom inflammation had already begun to subside, particularly those who received early immunosuppressive therapy, and a longer threshold would further risk overlap with chronic vascular changes. This trade‐off between capturing sufficient cases and maintaining specificity for active inflammation represents an inherent limitation of the study design. An important practical implication of our findings is that VWI detected arterial wall pathology in a substantial number of patients in whom conventional MRA was unremarkable. MRA abnormalities were present in only 5 patients (12.8%), whereas VWI revealed vessel wall involvement in 16 patients (41.0%). This discrepancy emphasizes the limitation of luminal imaging techniques, which can only detect disease when it has progressed to the point of luminal narrowing or occlusion [26, 33]. VWI, by directly visualizing the vessel wall, can identify early or isolated wall pathology in the pre‐stenotic phase [26, 33]. The diagnostic yield of conventional MRA may be significantly enhanced when combined with VWI. However, in the absence of a control group, the disease‐specific nature of these findings remains uncertain. From a clinical perspective, VWI may provide complementary information beyond conventional luminal imaging, particularly in detecting early or subclinical vessel wall involvement. This may contribute to a more comprehensive assessment of disease activity, especially in cases where conventional imaging appears unremarkable. In addition, VWI could potentially be useful in monitoring disease activity and treatment response over time, although this requires further validation in longitudinal studies. Limitations This study has several limitations. The sample size of 39 patients, while representing one of the largest VWI cohorts in NBS to date in the context of this rare disease, still limits statistical power, particularly for subgroup analyses. In addition, as this study was conducted in a tertiary referral center, the patient population may represent a more selected and clinically heterogeneous population. Statistically significant associations should be interpreted as hypothesis‐generating rather than confirmatory. The allocation of patients to attack and attack‐free groups was not prospective. This introduces selection bias, as these patients may represent a clinically more severe phenotype. The significant gender imbalance between groups further compounds this potential confounding. Similarly, we studied only a small number of patients during an attack with a significant gender imbalance. The results of this study may have been influenced by different immunosuppressive therapies and differential time intervals between NBS diagnosis and imaging. The 3‐month attack window may have included patients in whom inflammation had already begun to resolve. The findings were not confirmed by histopathological analysis; thus, vessel wall changes cannot be conclusively attributed to vasculitis. The absence of a control group further limits the interpretability of our findings, particularly regarding the specificity of the observed VWI abnormalities for NBS, as similar vessel wall changes may also occur in other inflammatory or nonspecific vascular conditions. In addition, inter‐reader agreement was not formally assessed, as discrepancies between readers were resolved by consensus. However, the systematic evaluation of 30 segments per patient using a standardized 3 Tesla VWI protocol, independent image assessment by two experienced neuroradiologists who were blinded to detailed clinical information, and exclusion of patients with confounding vascular risk factors strengthen the internal validity of the observations. Finally, no formal correction for multiple comparisons was applied due to the exploratory nature of the study, and the findings should therefore be interpreted with caution. Conclusions This study suggests that intracranial arterial wall involvement may be detectable by VWI in a substantial proportion of NBS patients and may represent a potentially underrecognized component of the disease spectrum. The predominance of findings in the posterior circulation and their higher frequency during the attack period suggest a potential association with inflammatory activity. Although venular inflammation remains the central and well‐established mechanism in NBS, these findings support the concept that arterial wall involvement may also contribute to the vascular heterogeneity of the disease. Whether venous and arterial processes occur independently or represent different manifestations of the same inflammatory mechanism across vessel types remains to be determined. Larger, prospective, and longitudinal studies are needed to clarify the pathophysiological significance and potential prognostic implications of these observations. Author Contributions Nursena Erener: data curation. Esra Kochan Kizilkilic: conceptualization, investigation, data curation, methodology, writing – review and editing, writing – original draft. Emire Seyahi: writing – review and editing, supervision. Aksel Siva: data curation. Abdullah Arcan: data curation. Osman Kizilkilic: conceptualization, writing – review and editing, supervision, methodology. Ugur Uygunoglu: writing – review and editing, supervision, methodology, conceptualization. Bora Korkmazer: data curation, formal analysis. Serdar Arslan: data curation. Ahmet Kursat Karaman: data curation. Funding The authors have nothing to report. Ethics Statement The study was approved by the Local Ethics Committee (Ethics ID: 960814, 2024). All volunteers were provided with information about the procedures. Written informed consent was obtained from all participants before the study. The study was performed according to the principles of the Declaration of Helsinki. Consent Informed consent was obtained from all subjects involved in the study. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Table S1: Demographic and clinical characteristics of patients imaged during the attack and attack‐free periods. Table S2: MRA and Vessel Wall Imaging Findings in Patients with MRA Abnormalities.