Vascular, inflammatory and immune factors associated with acute ischemic events in Takayasu arteritis: a 10-year retrospective matched case-control study Yang Liu, Ying Liu, Haizhuan An, Qian Li, Sumiao Liu, Ke Xu Abstract ObjectivesAcute ischemic events (AIE) constitute life-threatening complications of Takayasu arteritis (TAK). This study aimed to identify vascular, inflammatory and immune factors associated with concurrent AIE in TAK patients.MethodsWe conducted a 10-year retrospective 1:2 matched case-control study including TAK patients without exposure to glucocorticoids, immunosuppressants, or antithrombotic agents for at least 3 months. Clinical characteristics, disease activity, inflammatory markers, and lymphocyte subsets were compared between patients with and without AIE. Three sequential multivariable logistic regression models were constructed to identify variables independently associated with AIE. Model discrimination was evaluated using receiver operating characteristic (ROC) analysis with 1000-bootstrap internal validation, while calibration and decision curve analysis (DCA) were performed as exploratory within-sample assessments.ResultsThirty-seven patients with AIE and 74 matched TAK control without AIE were included. The mean age at TAK onset in the AIE group was 34.92 ± 10.83 years (range: 16–58 years). Conventional markers including ESR, CRP and homocysteine showed no independent correlation with AIE. Patients with AIE presented substantially higher frequencies of coronary, anterior cerebral, middle cerebral, and femoral artery stenosis, alongside significantly elevated ITAS-2010 scores, serum IgA and IL-6 levels, and peripheral Th cell counts (all p < 0.05). Severe long-term functional sequelae were also more frequent in the AIE group. Multivariate analysis confirmed that stenosis at these four arterial sites, together with elevated ITAS-2010 score, IgA, IL-6, and Th cell counts, were independently associated with AIE. The fully integrated vascular-inflammation-immunity model yielded an apparent AUC of 0.942 (95%CI 0.898-0.985) and a bootstrap optimism-corrected AUC of 0.908 (95%CI 0.843-0.959), with a corrected Brier score of 0.1191. Calibration and DCA supported satisfactory within-sample model performance but should be interpreted as exploratory because of the matched case-control design.ConclusionsSpecific vascular stenosis, high disease activity, elevated serum IgA and IL-6 levels, and increased peripheral Th cell counts are independently associated with AIE in TAK. These findings provide additional insight into the vascular, inflammatory, and immune characteristics accompanying ischemic complications in TAK and warrant confirmation in prospective multicenter studies. Introduction Takayasu arteritis (TAK) is a rare, chronic granulomatous vasculitis predominantly affecting the aorta and its major branches, with a marked predilection for women under 40 years of age (, ). The inflammatory cascade originates in the adventitia and progressively involves all layers of the arterial wall, leading to concentric thickening that may result in structural abnormalities such as stenosis, occlusion, dilatation, or aneurysm formation (). A well-documented hypercoagulable state in TAK further amplifies the thrombosis risk in vessels with preexisting lesions, often culminating in acute ischemic events (AIE). These events may involve multiple vascular territories, including the cerebral, coronary, mesenteric, renal, and peripheral arteries, and may lead to devastating outcomes such as stroke, myocardial infarction, intestinal ischemia, limb gangrene, and death (). Notably, AIE may still occur despite immunosuppressive therapy, highlighting the need for a better understanding of the vascular, inflammatory, and immune characteristics accompanying ischemic events in TAK. The pathogenesis of TAK-associated AIE is complex and multifactorial. Traditional cardiovascular risk factors such as dyslipidemia, fail to fully explain the high thrombosis burden in this population (), indicating that disease-specific mechanisms play an important role. Cumulative evidences have shown that endothelial dysfunction, systemic inflammation are closely linked to the prothrombotic milieu in TAK (). Routine inflammatory markers (–) including erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) and neutrophil-to-lymphocyte ratio (NLR), are widely used to assess disease activity in clinical practice; however, their relationship with ischemic manifestations in TAK remains incompletely understood. Previous studies have confirmed that plasma cells and serum immunoglobulin levels are associated with disease activity and treatment response (, ). As a core proinflammatory cytokine, Interleukin-6 (IL-6) also correlates with inflammatory activity in TAK (). In addition, T lymphocytes play central roles in granulomatous arterial inflammation (). Despite these advances, it remains unclear which vascular, inflammatory and immune parameters are independently associated with concurrent AIE in TAK patients. Several assessment tools have been established for evaluating disease activity in TAK, but each has limitations. The National Institutes of Health criteria relies partly on invasive angiography and have low accuracy for judging pathologically activity (). The Disease Extent Index for TAK does not integrate acute-phase reactants with imaging findings and demonstrates suboptimal agreement with physician global assessment (, ). The widely used Indian Takayasu Clinical Activity Score (ITAS)-2010 and its derivative ITAS-A () primarily assess clinical disease activity but do not adequately capture vascular structural damage. Imaging-based scoring systems, such as the PET Vascular Activity Score, provide valuable information regarding vascular inflammation but remain constrained by cost, radiation exposure, and limited accessibility (). Importantly, these assessment systems were not developed to comprehensively characterize the vascular, inflammatory and immune features associated with AIE, as their primary focus is overall disease activity and vascular progression rather than ischemic manifestations. Although comprehensive vascular imaging is recommended for newly diagnosed TAK patients, occult arterial stenosis and subclinical ischemic lesions may remain undetectable before symptomatic ischemic events occur, indicating that imaging findings alone may not fully reflect the biological characteristics accompanying AIE (). Although relevant progress has been made in exploring the pathophysiology and biomarkers of TAK, few studies have systematically evaluated the vascular involvement together with inflammatory and immune parameters in patients with AIE. Therefore, we conducted a retrospective 1:2 matched case-control study to compare vascular characteristics, disease activity, inflammatory markers, immunoglobulins, and lymphocyte subsets between hospitalized TAK patients with and without AIE. We further performed multivariable logistic regression analyses to identify vascular, inflammatory, and immune variables independently associated with concurrent AIE and evaluated the robustness of these cross-sectional associations using prespecified sensitivity analyses. Given the retrospective matched case-control design and the availability of biomarker measurements after or shortly following AIE onset, the present study was intended to characterize exploratory cross-sectional associations rather than to develop a model for prospective prediction of future ischemic events. Materials and methods This retrospective 1:2 matched case-control study was conducted at Shanxi Bethune Hospital, enrolling patients diagnosed with TAK who were admitted between January 2014 and December 2024. All participants met either the 1990 American College of Rheumatology (ACR) or the 2022 ACR/European Alliance of Associations for Rheumatology (EULAR) classification criteria for TAK (, ). A key inclusion criterion required all patients to be free from interfering medications for at least 3 months before enrollment, to eliminate drug effects on inflammatory and immune biomarkers. Eligible patients fell into two categories: (1) newly diagnosed, treatment-naive patients without prior use of TAK-related therapies or antithrombotic agents; (2) previously diagnosed patients who had voluntarily or medically discontinued glucocorticoids, immunosuppressants and coagulation-modifying drugs for ≥3 months. Cases included 37 TAK patients who developed AIE during hospitalization. AIE were defined as acute vascular events occurring in the arterial circulation, confirmed by imaging and resulting in tissue infarction or ischemia, attributable to in situ arterial thrombosis rather than embolic or atherosclerotic mechanisms. Eligible AIE subtypes consisted of acute ischemic stroke with radiological cerebral infarction, acute myocardial infarction, peripheral arterial thrombosis, and visceral arterial thrombosis. Pulmonary infarction was only included imaging (contrast-enhanced CT, CTA or PET-CT) confirmed pulmonary artery wall thickening, stenosis or occlusion attributable to TAK. Patients with AIE attributable to alternative etiologies were excluded, including: (1) pulmonary infarction secondary to venous thromboembolism without evidence of pulmonary vasculitis; (2) suspected ischemic events without clinical or radiological confirmation; (3) embolic events related to atrial fibrillation, antiphospholipid syndrome or other non-inflammatory conditions; (4) ischemia caused by cardiogenic embolism, non-TAK vascular stenosis or trauma. Controls were 74 TAK patients without AIE, matched to case at a 1:2 ratio by age, sex, and body mass index (BMI). All controls were recruited during the same study period and also satisfied the 3-month medication-free requirement. Additional exclusion criteria for all participants were as follows: (1) coexisting other connective tissue diseases or overlap syndrome; (2) history of malignant tumors, active infections, hematological disorders, severe hepatic or renal insufficiency, pregnancy, or lactation; (3) incomplete medical records; or (4) missing baseline immune and inflammatory biomarker data. This study was conducted in strict compliance with the ethical principles outlined in the Declaration of Helsinki. The study protocol was reviewed and approved by the Ethics Committee of Shanxi Bethune Hospital (Approval No.: LYLL-2025-004/PJ46). Specimen collection and laboratory analysis All clinical data were independently reviewed and extracted by two experienced rheumatologists. Fasting peripheral blood samples were collected on the next morning after hospital admission following a standardized operating protocol. All baseline samples were obtained before initiation of any new TAK-related medications during hospitalization. Emergency anticoagulant treatment for acute AIE was administered after sample collection and thus did not interfere with baseline biomarker results. Flow cytometry was utilized to quantify the total number of lymphocytes in each of the subpopulations (CD3+ T, CD4+ T, CD8+ T, B and natural killer (NK) cells) isolated from PB samples. Blood cells were collected using a BD FACSCalibur platform (BD Biosciences, Franklin Lakes, NJ, USA) and quantified using MultiSET software (BD Biosciences) after 50 μL of EDTA-anticoagulated venous blood was placed into A and B Trucount tubes (BD Biosciences). The CD3+, CD4+, and CD8+ T cell subsets were identified using the conjugated monoclonal antibodies fluorescein isothiocyanate (FITC)-CD3, allophycocyanin (APC)-CD4, peridinin chlorophyll protein (PerpCP)-CD45, and phycoerythrin (PE)-CD8 ((BD Biosciences). For B and NK cell populations, monoclonal antibodies FITC-CD3, APC-CD19, PerpCP-CD45, and PE-CD16+CD56 were used (BD Biosciences). Subsequently 20 μl of above-mentioned antibodies were added into A and B tubes, respectively, followed by vortex stirring and incubation at room temperature for 15 minutes. Finally, cells were analyzed by flow cytometry and quantified using MultiSET software. Within one hour of collection, blood samples from all TAK patients were centrifuged at 3000g for twenty minutes; the resulting serum was then used for cytokine assays. Using a Cytometric Bead Array, the concentrations of IL-2, IL-4, IL-6, IL-10, IL-17, INF-γ and TNF-α in the serum were determined. Every procedure followed the manufacturer’s instructions. Disease activity was assessed using the ITAS-2010 (). The Takayasu damage score (TADS) () was applied to evaluate TAK organ damage. Discrepancies between the two reviewers were resolved through consensus discussion. Any disagreements between the two assessors were resolved by group discussion to reach a consensus. Statistical analysis All statistical analyses were performed using SPSS 26.0 (IBM Corp., Armonk, NY, USA) and R software (version 4.5.3; R Foundation for Statistical Computing, Vienna, Austria). A two-sided p value <0.05 was considered statistically significant. Continuous variables were assessed for normality using the Shapiro-Wilk test. Normally distributed variables were expressed as mean ± standard deviation (SD) and compared using the independent-samples T test. Non-normally distributed variables were summarized as median (interquartile range, IQR) and compared using the Mann-Whitney U test. Categorical variables were presented as frequencies and percentages and compared using the chi-square test or Fisher’s exact test, as appropriate. Univariate logistic regression analysis was initially performed to identify variables associated with concurrent AIE. Variables with p-value < 0.05 in univariate analysis were entered into multivariable logistic regression analyses. Three hierarchical multivariate unconditional logistic regression models were constructed by biological dimension: Model 1 included vascular lesion variables only; Model 2 additionally incorporated inflammatory markers; and Model 3 further included peripheral T-lymphocyte subsets. A prespecified parsimonious sensitivity analysis was performed using biologically relevant vascular, inflammatory, and immune variables to evaluate the robustness of the observed cross-sectional associations and partially address potential model over-parameterization resulting from the limited number of AIE events. Because biomarkers in patients with AIE were measured after or shortly following the ischemic event, all multivariable analyses were intended to identify variables independently associated with concurrent AIE rather than to develop a model for prospective prediction of future ischemic events. The discriminatory performance of the sequential multivariable association models was evaluated using receiver operating characteristic (ROC) curves and the corresponding area under the curve (AUC). Internal validation was performed using 1000 bootstrap resamples to assess the stability of model discrimination. The apparent AUC, optimism-corrected AUC, and corresponding 95% confidence intervals (CIs) were calculated, and AUCs were compared using the DeLong test. Calibration curves and the Hosmer-Lemeshow goodness-of-fit test were generated as exploratory within-sample assessments of agreement between fitted probabilities and observed outcomes in this matched case-control dataset. Decision curve analysis (DCA) was likewise performed as an exploratory within-sample analysis to compare the relative statistical performance of the sequential models. Owing to the retrospective matched case-control design and post-event biomarker measurements, the calibration and DCA results were not intended to estimate absolute risk, demonstrate clinical utility, or support clinical decision-making. Results Clinical features and vascular involvement of TAK patients with and without AIE Details of the individual cases are provided in , . A total of 111 patients with TAK were included in this study, comprising 37 patients with AIE and 74 without AIE who served as the control group. The mean age at onset of TAK in the AIE group was 34.92 ± 10.83 years, ranging from 16 to 58 years. No statistically significant differences were found between the AIE and control groups in terms of gender (female: 83.3% vs 90.5%, p = 0.299) or BMI (22.60 ± 2.76 kg/m2 vs 23.18 ± 3.30 kg/m2, p = 0.368). The prevalence of smoking or alcohol consumption showed no significant difference between the AIE and control groups. Similarly, no significant differences were observed in the history of hypertension, diabetes or dyslipidemia between the AIE and control groups. The majority of patients with AIE were females, with only 6 male patients, consistent with the known sex distribution of TAK. As classified by Numano’s angiographic criteria (), type I and type V were the dominant patterns in the AIE group (12/37, 32.4%). Conversely, type V predominated in patients without AIE (37/74, 50%), followed by type I (22/74, 29.7%). The remaining subtypes were distributed as follows: type IV was observed in 4 patients (4 in the AIE group and 0 in the non-AIE group), type III in 6 (1 and 5, respectively), type IIa in 12 (5 and 7, respectively), and type IIb in 4 (1 and 3, respectively).We further compared vascular involvement between patients with and without AIE (, ). Compared with controls, patients with concurrent AIE exhibited significantly higher frequencies of anterior cerebral artery (ACA) stenosis (p = 0.007), middle cerebral artery (MCA) stenosis (p = 0.001), coronary artery stenosis (p = 0.001) and femoral artery stenosis (p = 0.028). No significant differences were observed for the remaining cervicocephalic, aortic, or visceral branch arteries (all p > 0.05), although pulmonary and iliac artery involvement showed borderline between-group differences (both p = 0.062). The first episode of AIE occurred from the time of initial presentation to over 28 years after the onset of TAK, with a mean duration of TAK of 77.68 ± 108.89 months before AIE onset. The interval between the first disease manifestation and definitive TAK diagnosis averaged 39.14 ± 84.56 months, whereas the mean interval between TAK diagnosis and AIE occurrence was 38.56 ± 72.20 months. Among the 37 patients with AIE, 23 were identified at the time of diagnosis, 1 occurred 4 months before TAK diagnosis, and the rest developed during follow-up at 1 (n=7), 3 (n=1), 5 (n=2), 10 (n=1), and 20 years(n=2) after diagnosis (). The anatomical distribution of thrombotic events in TAK is illustrated in , with cerebral artery thrombosis (stroke) being the most common ischemic manifestation. The median follow-up duration was comparable between the two groups (56.64 ± 35.72 vs 62.53 ± 55.63 months, p = 0.564). Patients with AIE experienced a numerically higher number of disease relapses than controls, although the difference did not reach statistical significance (1.86 ± 2.08 vs 1.26 ± 1.60, p = 0.091). As summarized in , , severe long-term functional sequelae were substantially more frequent among patients with AIE. Several patients also developed major adverse events (MAEs) during follow-up. Notably, one patient developed pyoderma gangrenosum 14 months after TAK diagnosis and later experienced extensive embolic occlusion involving the anterior and posterior tibial, peroneal, femoral, and popliteal arteries at 50 months post-diagnosis, ultimately requiring a right transfemoral amputation (). This case illustrates the heterogeneous clinical course of TAK and highlights that extensive vascular involvement may occur even several years after diagnosis. Inflammation markers, immunity parameters and disease activity in patients with and without concurrent AIE As summarized in , patients with concurrent AIE exhibited more extensive disease involvement. Seizures were significantly more frequent in the AIE group, and the number of blocked arteries was also significantly higher than that in patients without AIE (both p < 0.05). No significant between-group differences were observed in other clinical manifestations, including fever, fatigue, cardiovascular involvement, or other complications. For disease activity and vascular damage, The ITAS-2010 score was markedly elevated in patients with AIE than in controls (7.46 ± 3.08 vs 5.49 ± 2.56, p < 0.001). The AIE group also exhibited a modestly higher TADS (4.00 [2.00, 5.00] vs 3.00 [2.00, 5.00], p = 0.041), indicating greater cumulative vascular damage. Among inflammatory markers, serum IgA levels were significantly higher in the AIE group (3.19 [2.24, 4.98] g/L vs 2.42(1.76,3.62) g/L, p = 0.013). Serum IL-6 concentrations were also remarkably increased in the AIE group (30.50 [8.60, 49.23] pg/mL vs 4.27(2.68,12.12) pg/mL, p < 0.001), suggesting a more pronounced inflammatory profile in patients presenting with concurrent AIE. Analysis of lymphocyte subsets revealed significant between-group differences. Th cell counts were significantly higher in the AIE group (1124.00 [851.50, 1546.50] cells/μL vs 853.50[586.75,1053.00] cells/μL, p=<0.001). Similarly, Ts cell levels were also markedly elevated in patients with AIE (715.00 [507.50, 984.50] cells/μL vs 487.00[336.50,667.25] cells/μL, p < 0.001). No significant differences were observed in B cells, NK cells, or other circulating cytokines between the two groups (all p>0.05). As shown in , autoantibody profiles were comparable between groups. Multivariable analysis of factors associated with concurrent AIE in patients with TAK Based on the aforementioned findings, multivariate logistic regression analyses were performed to identify vascular, inflammatory, and immune variables independently associated with concurrent AIE in TAK, and three sequential hierarchical models were constructed (). Model 1 included vascular involvement only. Coronary artery, MCA and femoral artery were independently associated with concurrent AIE (all p < 0.05), whereas ACA stenosis showed a borderline association (p = 0.080). The total number of blocked arteries was not independently associated with concurrent AIE (p = 0.837). Model 2 incorporated inflammatory variables (ITAS-2010, IgA and IL-6) in addition to vascular lesions. Coronary, ACA, and MCA stenosis remained independently associated with concurrent AIE (all p < 0.05), whereas the association for femoral artery stenosis became borderline significant (p = 0.064). Higher ITAS-2010 score, serum IgA and IL-6 levels also independently associated with concurrent AIE (all p < 0.05). Model 3 further incorporated peripheral T-lymphocyte subsets. In this fully adjusted model, coronary artery stenosis (OR = 18.958, 95% CI: 1.546-232.548, p = 0.021), ACA stenosis (OR = 9.450, 95% CI: 1.429-62.480, p = 0.020), MCA stenosis (OR = 5.948, 95% CI: 1.179-30.019, p = 0.031), femoral artery stenosis (OR = 35.026, 95% CI: 1.409-870.693, p = 0.030), higher ITAS-2010 score (OR = 1.327, 95% CI: 1.038-1.696, p = 0.024), elevated serum IgA (OR = 1.830, 95% CI: 1.056-3.171, p = 0.031), increased IL-6 levels (OR = 1.056, 95% CI: 1.022-1.091, p = 0.001), and increased Th cell count (OR = 1.003, 95% CI: 1.001-1.005, p = 0.009) remained independently associated with concurrent AIE. To evaluate the robustness of these findings, a prespecified parsimonious sensitivity analysis was performed. The direction and magnitude of the associations remained largely consistent with those observed in the primary model. Although the association for coronary artery stenosis was attenuated and became borderline significant (OR = 6.54, 95% CI: 0.87-48.92, p = 0.067), the direction of effect remained unchanged. The associations for MCA stenosis, ITAS-2010 score, serum IgA, IL-6, and Th cell count remained stable across the primary and sensitivity analyses (), supporting the robustness of the principal cross-sectional associations. Discriminatory performance, internal validation, and exploratory model evaluation The discriminatory performance of the three sequential multivariate association regression models was evaluated using ROC curve analysis, 1000-bootstrap internal validation, exploratory calibration curves, and DCA. As illustrated in the ROC plot (), the apparent AUC values increased progressively across the three models: Model 1 yielded an apparent AUC of 0.824 (95% CI: 0.741-0.907), Model 2 reached 0.913 (95% CI: 0.860-0.967), and the fully adjusted Model 3 attained the highest apparent AUC of 0.942 (95% CI: 0.898-0.985). After 1000-bootstrap optimism correction, the corrected AUCs remained consistent with the apparent values: 0.794 (95% CI: 0.699-0.880) for Model 1, 0.878 (95% CI: 0.816-0.936) for Model 2, and 0.908 (95% CI: 0.843-0.959) for Model 3. Similar improvements were observed for the Brier score. The apparent Brier scores were 0.1548, 0.1104, and 0.0846 for Model 1-3, respectively, whereas the optimism-corrected Brier scores were 0.1773, 0.1423, and 0.1191, indicating improved within-sample discrimination and overall statistical fit with the sequential incorporation of inflammatory and immune variables. Pairwise comparisons of AUCs using the DeLong test demonstrated significant differences between Model 1 and Model 2 (AUC difference = -0.090, z = -2.645, p = 0.008), as well as between Model 1 and Model 3 (AUC difference = -0.118, z = -3.432, p = 0.001). Although the AUC difference between Model 2 and Model 3 did not reach statistical significance (AUC difference = -0.028, z = -1.770, p = 0.077), a likelihood ratio test demonstrated that adding Th and Ts cell counts significantly improved model fit (χ2=15.86, df = 2, p < 0.001). Calibration analysis () was performed as an exploratory within-sample assessment of agreement between fitted probabilities and observed outcomes in this matched case-control dataset. The mean absolute calibration error was 0.066 for Model 1, 0.049 for Model 2, 0.057 for Model 3. Among the three models, Model 3 demonstrated the closest agreement between fitted probabilities and observed outcomes, with its bias-corrected calibration curve closely approximating the ideal reference line and the 95% confidence band encompassing the reference diagonal. Decision curve analysis () was likewise performed as an exploratory within-sample comparison of the relative statistical performance of the three sequential models. Owing to the retrospective matched case-control design and the availability of biomarker measurements after or shortly following AIE onset, these analyses were not intended to estimate absolute risk, demonstrate clinical utility, or support prospective clinical decision-making. Across threshold probabilities of approximately 0.05-0.80, Model 3 exhibited the highest net benefit among the three models, whereas the net benefit of all models gradually declined at threshold probabilities exceeding 0.80. Overall, these findings indicate that integrating vascular, inflammatory, and immune variables improved the within-sample discriminatory performance of the association models; however, these results should be interpreted as exploratory statistical summaries rather than evidence supporting prospective risk stratification or prediction of future AIE. Discussion AIE represent one of the most devastating complications of TAK and substantially contribute to long-term disability and mortality. In this 10-year single-center retrospective matched case-control study, we systematically evaluated vascular, inflammatory, and immune characteristics in patients with and without concurrent AIE. We found that stenosis involving the coronary arteries, ACA, MCA, and femoral arteries, together with higher ITAS-2010 scores, elevated serum IgA and IL-6 levels, and increased peripheral Th cell counts, were independently associated with concurrent AIE after multivariable adjustment. Furthermore, sequential multivariable association models suggested that integrating vascular, inflammatory, and immune variables provided a more comprehensive characterization of the biological profile observed in patients presenting with AIE than vascular features alone. These findings extend previous observations regarding the immune-inflammatory basis of ischemic complications in TAK. Accordingly, these findings should be regarded as exploratory associations that warrant confirmation in independent prospective cohorts. Multiple ethnic cohort studies (–) have consistently demonstrated that ischemic complications in TAK are closely related to the extent and distribution of vascular involvement. In the present study, Numano type I and type V angiographic patterns predominated among patients with AIE, whereas isolated type V disease was most common in patients without AIE. More importantly, stenosis involving ACA, MCA, coronary artery, and femoral artery remained independently associated with AIE after multivariable adjustment, whereas the total number of occluded arteries showed no independent association. These findings suggest that the anatomical distribution of vascular lesions may be more informative than overall vascular burden in characterizing ischemic complications in TAK. A previous meta-analysis () reported that stroke and myocardial infarction occur in approximately 8.9% and 3.4% of patients with TAK, respectively, underscoring the substantial burden of ischemic complications in this disease. The predominance of cerebrovascular events in our cohort is consistent with these observations and is biologically plausible. Consistent with our findings, a 320-patient TAK cohort reported cerebrovascular ischemia occurred in 20% of patients, frequently accompanied by multivessel stenosis (). Similarly, a long-term survival study including 224 patients identified stroke and myocardial infarction as the leading cause of death in TAK, with mortality substantially exceeding that of the age- and sex-matched general populations (SMR = 17.29) (). Femoral artery involvement may reflect extensive systemic vasculitis rather than isolated peripheral arterial disease. Supporting this interpretation, an FDG-PET/CT study of 77 patients found that femoral artery inflammation was associated with older age, male sex, and a higher burden of atherosclerotic risk factors (). Patients with AIE exhibited higher ITAS-2010 scores, more severe vascular occlusion, and higher TADS, supporting that sustained vascular inflammation and cumulative structural damage are closely linked to ischemic complications in TAK (). Conventional inflammatory markers (ESR, CRP and NLR) showed no independent correlation with AIE, consistent with prior work indicating these routine indicators poorly reflect long-term thrombotic predisposition (–). We also evaluated traditional cardiovascular risk factors potentially related to arterial thrombosis. Although serum homocysteine levels were slightly elevated in the AIE group, no significant intergroup difference emerged, consistent with previous studies (, ). Collectively, these findings suggest that conventional cardiovascular risk factors and routinely used inflammatory biomarkers have limited ability to capture the heterogeneous immune-inflammatory status accompanying ischemic complications in TAK. In contrast, serum IgA and IL-6 levels were significantly higher in patients with AIE and remained independently associated with AIE after multivariable adjustment. Rather than reflecting nonspecific systemic inflammation alone, these biomarkers may better characterize the immune-inflammatory milieu accompanying ischemic events in TAK. Their independent associations observed in the present study further support the concept that immune activation represents an important component of the biological characteristics of patients presenting with AIE and provides a rationale for investigating these pathways in future mechanistic and prospective studies. The vascular inflammation characteristic of TAK is driven by complex interactions between humoral and cellular immunity. In the present study, serum IgA, IL-6, and peripheral Th cell counts were all independently associated with concurrent AIE after multivariable adjustment, supporting the involvement of adaptive immune activation in patients presenting with ischemic complications. Although the cross-sectional design precludes causal inference, these findings extend previous observations by demonstrating that both humoral and cellular immune responses are closely linked to the immune-inflammatory characteristics accompanying AIE. Serum IgA was significantly elevated in patients with AIE and was verified as an independent AIE-associated factor after adjusting for confounding variables. Previous studies have suggested that increased IgA may reflect persistent mucosal or systemic immune activation, and IgA-mediated immune complexes can promote endothelial injury, platelet aggregation, and arterial thrombosis (, , ). Our findings expand the potential clinical relevance of IgA in TAK by suggesting that, beyond its reported association with disease activity, elevated IgA also characterizes patients presenting with ischemic complications. IL-6 is a pivotal cytokine in the inflammatory network of TAK and has consistently been associated with disease activity and vascular inflammation. In our cohort, IL-6 showed a prominent correlation with AIE and was markedly higher in patients with ischemic complications. Cumulative preclinical and clinical evidence (, ) confirms excessive IL-6 signaling promotes endothelial dysfunction, enhances acute-phase responses, and disrupt the balance between coagulation and fibrinolysis, thereby contributing to a prothrombotic vascular environment. The present findings further support the close relationship between IL-6-mediated immune activation and ischemic manifestations in TAK. Abnormal T-cell activation represents another hallmark of TAK pathogenesis (). Peripheral Th cell counts were independently associated with AIE, whereas Ts cell counts were no longer significant after adjustment for other variables. CD4+ Th cells are central mediators of granulomatous arterial inflammation and contribute to vascular remodeling through the production of pro-inflammatory cytokines, including IL-6 and IFN-γ (, ). Moreover, activated T lymphocytes participate in the crosstalk between inflammation and coagulation within the injured vascular wall (). Although the present findings cannot establish temporal or causal relationships, they support the concept that adaptive immune activation constitutes an important biological feature of TAK patients presenting with concurrent AIE. To better evaluate the relative contributions of vascular, inflammatory, and immune variables, we constructed three sequential multivariate association models. The stepwise incorporation of inflammatory biomarkers and T-lymphocyte subsets progressively improved within-sample discrimination, while likelihood ratio testing demonstrated that the addition of T-cell subsets significantly improved overall model fit. Although the difference in AUC between Models 2 and 3 did not reach statistical significance, the bootstrap-corrected discrimination remained stable, suggesting that immune variables contributed additional information beyond vascular and inflammatory characteristics alone. Given the relatively limited number of AIE events, we further performed a prespecified parsimonious sensitivity analysis to address potential model over-parameterization. The principal associations for MCA stenosis, ITAS-2010 score, serum IgA, IL-6, and Th cell count remained largely unchanged after model simplification. Although the association for coronary artery stenosis became borderline significant in the sensitivity analysis, the direction and magnitude of the association were generally preserved, suggesting that the principal findings were reasonably robust despite the limited sample size. Nevertheless, the wide confidence intervals observed for several vascular variables indicate limited precision of the regression estimates and should be interpreted with appropriate caution. To provide a comprehensive statistical evaluation, we additionally performed bootstrap internal validation, calibration analysis, and DCA. These analyses consistently supported the overall statistical performance of the sequential association models within the study sample. However, because this was a retrospective matched case-control study with biomarker measurements obtained after or shortly following AIE onset, the calibration and decision curve analyses should be regarded as exploratory assessments of within-sample model performance rather than evidence of prospective prediction, absolute risk estimation, or clinical decision-making utility. External validation in independent prospective multicenter cohorts remains essential before any broader application of these statistical models can be considered. The present findings provide additional insight into the vascular, inflammatory, and immune characteristics of TAK patients presenting with concurrent AIE. Rather than relying on conventional inflammatory markers alone, integrating information on vascular involvement, disease activity, humoral immunity, and T-cell activation may offer a more comprehensive understanding of the biological heterogeneity accompanying ischemic complications in TAK. These observations complement previous studies focusing predominantly on angiographic manifestations by highlighting the potential contribution of adaptive immune activation to ischemic phenotypes. From a clinical perspective, the identified vascular and immune-inflammatory variables may assist in the comprehensive assessment of patients presenting with AIE, although they should not be interpreted as validated predictors of future ischemic events. Instead, these findings generate hypotheses for future mechanistic investigations and prospective longitudinal studies designed to clarify the temporal relationship between immune activation and ischemic complications. Such studies will be essential to determine whether dynamic changes in these biomarkers precede AIE onset and whether they provide incremental value beyond conventional clinical and imaging assessments. Several limitations of this study should be acknowledged. First, this was a retrospective single-center matched case-control study, and fasting blood samples in patients with AIE were obtained at admission, mostly after or shortly following the ischemic event. Therefore, the temporal relationship between immune biomarkers and AIE cannot be established, and reverse causation cannot be excluded. Second, the enriched case-control design restricted to treatment-naïve patients without recent glucocorticoid, immunosuppressive, or antithrombotic therapy may have introduced selection bias and limits the generalizability of our findings to broader TAK populations. Third, the relatively small number of AIE events limited statistical power and increased the risk of model over-parameterization, as reflected by the wide confidence intervals for several vascular variables. Although a prespecified parsimonious sensitivity analysis demonstrated generally consistent associations, larger prospective cohorts using prespecified or penalized regression approaches are needed to confirm these findings. Fourth, the included ischemic events represented a heterogeneous group of clinical outcomes, and no prespecified prediction horizon was defined. Consequently, the multivariable models should be interpreted as exploratory analyses of cross-sectional associations rather than tools for prospective risk prediction. Finally, only internal bootstrap validation was performed, peripheral immune biomarkers were measured at a single time point, and detailed T-cell phenotyping was unavailable. External multicenter validation with longitudinal serial immune profiling will be essential to verify these observations and further clarify the mechanisms underlying ischemic complications in TAK. In conclusion, patients with TAK presenting with AIE exhibited higher disease activity, more extensive vascular involvement, and enhanced humoral and cellular immune activation. Stenosis of the coronary, anterior cerebral, middle cerebral, and femoral arteries, together with higher ITAS-2010 scores, elevated serum IgA and IL-6 levels, and increased peripheral Th cell counts, were independently associated with concurrent AIE. These findings expand current understanding of the vascular and immune-inflammatory characteristics associated with ischemic complications in TAK and provide a foundation for future prospective studies to determine their temporal and prognostic significance.