Association of Traditional and Novel Lipid Indicators With the Hemorrhagic Phenotype in Adult Moyamoya Disease: Implications for Lipid Risk Stratification

Association of Traditional and Novel Lipid Indicators With the Hemorrhagic Phenotype in Adult Moyamoya Disease: Implications for Lipid Risk Stratification | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Association of Traditional and Novel Lipid Indicators With the Hemorrhagic Phenotype in Adult Moyamoya Disease: Implications for Lipid Risk Stratification Weihong Huang, Haoyang Xiong, Wenjing Chen, Wei Liu, Peicong Ge, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9393575/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Objective To investigate the associations of traditional and novel lipid parameters with the hemorrhagic phenotype in adults with primary moyamoya disease (MMD) and to evaluate their potential value for phenotype discrimination and risk stratification. Methods In this retrospective dual-center cross-sectional study, 1,176 adults with primary MMD treated at Beijing Hospital and Beijing Tiantan Hospital between January 2022 and January 2026 were included, comprising 857 patients with a non-hemorrhagic phenotype and 319 with a hemorrhagic phenotype. Traditional and novel lipid parameters were derived from routine lipid measurements. Missing body mass index (BMI) values were handled using multiple imputation. Multivariable logistic regression, restricted cubic spline analysis, receiver operating characteristic analysis, incremental predictive analysis, sensitivity analyses, subgroup analyses, and mediation analysis were performed. Results Compared with patients with the non-hemorrhagic phenotype, those with the hemorrhagic phenotype had lower BMI and higher levels of multiple cholesterol-related lipid parameters. In multivariable analyses, total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and non-high-density lipoprotein cholesterol (non-HDL-C) were significantly associated with the hemorrhagic phenotype, with odds ratios for the highest versus lowest quartile of 3.435 (95% CI, 2.254–5.237), 3.197 (95% CI, 2.134–4.791), and 3.150 (95% CI, 2.087–4.754), respectively. TC, LDL-C, and non-HDL-C showed modest discriminative ability and improved the performance of the basic clinical model. Sensitivity and subgroup analyses were generally consistent. BMI showed a significant negative mediating effect in the associations between several key novel lipid parameters and the hemorrhagic phenotype. Conclusions Cholesterol-related traditional and novel lipid parameters, particularly TC, LDL-C, and non-HDL-C, were independently associated with the hemorrhagic phenotype in adult primary MMD and may provide clinically accessible markers for phenotype stratification. moyamoya disease hemorrhagic phenotype lipid parameters non-HDL cholesterol risk stratification Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Moyamoya disease (MMD) is a chronic cerebrovascular disorder characterized by progressive stenosis or occlusion of the terminal internal carotid arteries and the proximal anterior and middle cerebral arteries, accompanied by the formation of an abnormal collateral vascular network at the base of the brain.[ 1 , 2 ] In adults, stroke is the major clinical outcome, and the hemorrhagic phenotype is associated with higher disability and mortality. Its occurrence is closely related to fragile collateral circulation, abnormal hemodynamic burden, and dysregulated vascular wall remodeling[ 3 , 4 ]. In recent years, in addition to genetic susceptibility and perfusion abnormalities, lipid metabolic disturbances have also been suggested to participate in the development and progression of MMD[ 5 ]. Previous studies have shown that patients with MMD may exhibit abnormalities not only in traditional lipid indices but also in nontraditional lipid markers such as oxLDL, sdLDL, Lp(a), and non-HDL-C, some of which are associated with disease progression or the risk of cerebrovascular events[ 5 , 6 ]. These findings suggest that lipid abnormalities may contribute to the pathophysiology of MMD through oxidative stress, endothelial injury, inflammatory activation, and abnormal vascular remodeling[ 7 ]. However, existing evidence has mainly focused on the overall risk of MMD[ 8 ], postoperative recurrent stroke[ 6 ], or single metabolic factors[ 5 ]. Systematic studies examining the lipid metabolic profile reflected jointly by traditional and nontraditional lipid parameters in relation to hemorrhagic risk in adults with MMD, particularly the hemorrhagic phenotype, remain limited. Therefore, the present study jointly analyzed traditional and nontraditional lipid parameters to identify lipid metabolic features associated with the hemorrhagic phenotype in adults with MMD and to explore their potential value in hemorrhagic risk assessment and prediction. Methods Study Population This was a retrospective dual-center cross-sectional observational study. We consecutively screened 1,901 patients with MMD who were treated at Beijing Hospital and Beijing Tiantan Hospital, Capital Medical University, between January 2022 and January 2026, including 665 patients from Beijing Hospital and 1,236 from Beijing Tiantan Hospital. All patients were diagnosed with MMD based on clinical manifestations and cerebrovascular imaging findings. Moyamoya syndrome was excluded according to the criteria of the Research Committee on Spontaneous Occlusion of the Circle of Willis and the 2021 Japanese Guidelines for the Management of Moyamoya Disease[ 9 ], and therefore all included cases represented primary MMD. The inclusion criteria were as follows: (1) diagnosis of MMD based on digital subtraction angiography (DSA); (2) complete demographic, clinical, and imaging data; and (3) lipid testing performed within 24 h of admission, with complete measurements of total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C). The exclusion criteria were: (1) missing key lipid measurements (TC, TG, HDL-C, or LDL-C; n = 470); (2) age < 18 years (n = 243); and (3) incomplete clinical or imaging data that precluded subsequent statistical analysis (n = 12). Ultimately, 1,176 adults with MMD were included in the final analysis, comprising 857 patients with a non-hemorrhagic phenotype and 319 with a hemorrhagic phenotype (Fig. 1 ). Hemorrhagic MMD was defined as intracranial hemorrhage confirmed by cranial CT or MRI either at admission or previously, including intracerebral hemorrhage, intraventricular hemorrhage, or subarachnoid hemorrhage. Non-hemorrhagic MMD included asymptomatic cases and ischemic cases, including transient ischemic attack (TIA) and cerebral infarction. Data Collection Demographic, clinical, and laboratory data were collected for all included patients. Demographic and clinical variables included age, sex, body mass index (BMI), smoking history, drinking history, hypertension, diabetes mellitus, lesion laterality (unilateral or bilateral), and Suzuki stage. Patients were classified as having a hemorrhagic or non-hemorrhagic phenotype according to clinical presentation at admission and imaging findings. Laboratory data were obtained from routine hematologic and lipid tests performed at admission, including neutrophil count, lymphocyte count, monocyte count, platelet count, TG, TC, LDL-C, and HDL-C. Imaging data were independently reviewed in a blinded manner by two experienced neurosurgeons; when discrepancies occurred, a third physician adjudicated the final result. Exposure Definitions The exposures of interest consisted of baseline traditional lipid parameters and their derived nontraditional lipid parameters. Traditional lipid parameters included TC, TG, HDL-C, and LDL-C, all obtained from the first laboratory test performed after admission. Based on these measurements, the following nontraditional lipid parameters were calculated for subsequent analyses: atherogenic index of plasma (AIP), non-HDL-C, atherogenic coefficient (AC), Castelli's index-I (CRI-I), Castelli's index-II (CRI-II), lipoprotein combined index (LCI), remnant cholesterol (RC), and the RC/HDL-C ratio. The formulas were as follows: (1) AIP = lg(TG / HDL-C) [ 10 ] (2) non-HDL-C = TC - HDL-C [ 11 ] (3) AC = non-HDL-C / HDL-C [ 12 ] (4) CRI-I = TC / HDL-C [13] (5) CRI-II = LDL-C / HDL-C [ 14 ] (6) LCI = TC x TG x LDL-C / HDL-C [ 15 ] (7) RC = TC - LDL-C - HDL-C [ 16 ] (8) RC/HDL-C = RC / HDL-C [ 17 ] Statistical Analysis Continuous variables were first tested for normality using the Shapiro-Wilk test. Because all continuous variables were nonnormally distributed, they are presented as medians with interquartile ranges (IQRs), and between-group comparisons were performed using the Mann-Whitney U test. Categorical variables are presented as counts and percentages and were compared using the chi-square test or Fisher's exact test, as appropriate. Given the lack of established lipid cutoffs for patients with MMD, each lipid parameter was further categorized into quartiles for comparative analyses. BMI was the only variable with missing data, with 147 missing values among 1,176 patients (12.5%). Missing BMI values were handled using multiple imputation with the mice package, generating five imputed datasets. The imputation model included the outcome variable, lipid parameters, and relevant covariates. Logistic regression models were used to analyze the associations of traditional and nontraditional lipid parameters with hemorrhagic MMD, and odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. Three progressively adjusted models were constructed: Model 1 was unadjusted; Model 2 was adjusted for age, sex, and BMI; and Model 3 was additionally adjusted for hypertension, diabetes, smoking, drinking, bilateral involvement, and Suzuki stage. In quartile analyses, the lowest quartile (Q1) served as the reference group. To assess multicollinearity, variance inflation factors (VIFs) were calculated for independent variables in each model, and correlation matrices were generated to evaluate correlations among lipid parameters. A VIF < 5 was considered indicative of no significant multicollinearity (Supplementary Fig. 1). Restricted cubic spline (RCS) models were further fitted to assess dose-response relationships and potential nonlinear associations between key lipid parameters and hemorrhagic MMD, using four prespecified percentiles as knots and the median value as the reference. RCS analyses were performed with the rms package. To evaluate the robustness of the findings, sensitivity analyses were conducted, including repeated analyses in the complete-case dataset and repeated analyses after further dividing the non-hemorrhagic group into asymptomatic and ischemic subgroups (TIA and cerebral infarction). Additional subgroup analyses were performed according to age (stratified at the median age of 42 years), sex, hypertension status, and BMI ( = 28), and potential effect modification was tested by including multiplicative interaction terms in the main models. Receiver operating characteristic (ROC) curves and areas under the curve (AUCs) were used to evaluate the discriminative ability of traditional and nontraditional lipid parameters for hemorrhagic MMD, with internal validation by 100 bootstrap resamples. AUCs were compared pairwise using the DeLong test. The incremental predictive value of traditional and nontraditional lipid parameters was further assessed using the integrated discrimination improvement (IDI) and net reclassification improvement (NRI). ROC-related analyses were performed using the pROC package. In addition, mediation analysis was conducted using the Rmediation package to evaluate the potential mediating role of BMI in the associations between lipid parameters and hemorrhagic MMD[ 17 ]. After adjustment for covariates, indirect effects, direct effects, total effects, and the proportion mediated were estimated. All statistical tests were two-sided, and P < 0.05 was considered statistically significant. All analyses were performed using R 4.5.2. Results Clinical Characteristics of the MMD Cohort A total of 1,176 adults with primary MMD were included, including 857 patients with a non-hemorrhagic phenotype and 319 with a hemorrhagic phenotype. Among the non-hemorrhagic cases, 107 were asymptomatic and 750 were ischemic (including TIA and cerebral infarction). Compared with the non-hemorrhagic group, patients with the hemorrhagic phenotype had lower BMI and lower proportions of hypertension, diabetes, smoking, and bilateral involvement, and they also differed significantly in the distribution of Suzuki stage. By contrast, age, sex, and drinking status did not differ significantly between groups. In terms of lipid parameters, patients with the hemorrhagic phenotype had significantly higher levels of TC, LDL-C, HDL-C, non-HDL-C, AC, CRI-I, CRI-II, and LCI, whereas AIP was significantly lower. No significant between-group differences were observed for TG, RC, or RC/HDL-C (Table 1 ). Table 1 Baseline characteristics of patients with non-hemorrhagic and hemorrhagic phenotypes Variable Overall Non-hemorrhagic phenotype Hemorrhagic phenotype P value Demographic and clinical characteristics Sex, male, n (%) 522 (44.4%) 391 (45.6%) 131 (41.1%) 0.162 Age, years 42.00 (34.00, 50.00) 43.00 (34.00, 51.00) 42.00 (34.00, 50.00) 0.625 BMI, kg/m² 25.00 (22.84, 27.68) 25.38 (23.15, 28.03) 24.22 (21.86, 26.38) < 0.001 Hypertension, n (%) 448 (38.1%) 361 (42.1%) 87 (27.3%) < 0.001 Diabetes, n (%) 161 (13.7%) 144 (16.8%) 17 (5.3%) < 0.001 Smoking, n (%) 216 (18.4%) 170 (19.8%) 46 (14.4%) 0.033 Drinking, n (%) 150 (12.8%) 119 (13.9%) 31 (9.7%) 0.057 Bilateral involvement, n (%) 1035 (88.0%) 766 (89.4%) 269 (84.3%) 0.018 Suzuki stage, n (%) 0.006 ≤ 4 888 (75.5%) 629 (73.4%) 259 (81.2%) > 4 288 (24.5%) 228 (26.6%) 60 (18.8%) Lipid parameters TG, mmol/L 1.23 (0.85, 1.71) 1.25 (0.89, 1.74) 1.18 (0.81, 1.63) 0.064 TC, mmol/L 3.97 (3.32, 4.63) 3.83 (3.23, 4.50) 4.25 (3.68, 4.91) < 0.001 LDL-C, mmol/L 2.24 (1.68, 2.82) 2.14 (1.58, 2.65) 2.49 (1.96, 3.16) < 0.001 HDL-C, mmol/L 1.18 (1.00, 1.38) 1.16 (0.98, 1.35) 1.23 (1.04, 1.42) < 0.001 AIP 0.01 (-0.18, 0.21) 0.02 (-0.15, 0.22) -0.03 (-0.23, 0.19) 0.007 Non-HDL-C 2.74 (2.14, 3.38) 2.64 (2.05, 3.26) 2.99 (2.40, 3.61) < 0.001 Atherogenic coefficient 2.26 (1.72, 3.02) 2.23 (1.68, 2.95) 2.38 (1.87, 3.20) 0.002 CRI-I 3.26 (2.72, 4.02) 3.23 (2.68, 3.95) 3.38 (2.87, 4.20) 0.002 CRI-II 1.85 (1.39, 2.46) 1.82 (1.35, 2.39) 1.99 (1.52, 2.61) < 0.001 LCI 8.87 (4.63, 17.51) 8.61 (4.46, 16.28) 9.69 (5.27, 20.24) 0.017 RC 0.42 (0.30, 0.57) 0.42 (0.30, 0.57) 0.43 (0.30, 0.58) 0.491 RC/HDL-C 0.36 (0.24, 0.53) 0.36 (0.24, 0.53) 0.37 (0.23, 0.51) 0.564 Note: BMI, body mass index; TG, triglycerides; TC, total cholesterol; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; LCI, lipoprotein combined index; AIP, atherogenic index of plasma; AC, atherogenic coefficient; CRI-I, Castelli's index-I; CRI-II, Castelli's index-II; RC, remnant cholesterol. Associations of Traditional and Nontraditional Lipid Parameters with Hemorrhagic Risk in MMD After categorization of traditional and nontraditional lipid parameters into quartiles, logistic regression analyses were performed using the lowest quartile (Q1) as the reference to evaluate the association between the highest quartile (Q4) and hemorrhagic MMD. Multivariable logistic regression showed that, after progressive adjustment for age, sex, BMI, hypertension, diabetes, smoking, drinking, bilateral involvement, and Suzuki stage, TC and LDL-C among traditional lipid parameters, as well as non-HDL-C, AC, CRI-I, CRI-II, and LCI among nontraditional lipid parameters, were all significantly associated with hemorrhagic MMD (Fig. 2 ; Supplementary Table 1). Among them, TC, LDL-C, and non-HDL-C showed the strongest associations: the ORs for Q4 versus Q1 were 3.435 (95% CI: 2.254–5.237), 3.197 (95% CI: 2.134–4.791), and 3.150 (95% CI: 2.087–4.754), respectively, with all P for trend values < 0.001. In contrast, TG and HDL-C among traditional parameters and AIP, RC, and RC/HDL-C among nontraditional parameters were not significantly associated with hemorrhagic MMD after full adjustment. Overall, higher quartile levels of cholesterol-related traditional and nontraditional lipid parameters were closely associated with the hemorrhagic phenotype of MMD, and the strength of association increased across quartiles. Restricted Cubic Spline Analysis Multivariable-adjusted RCS analyses further showed significant overall associations of TC, LDL-C, non-HDL-C, AC, CRI-I, CRI-II, and LCI with the hemorrhagic phenotype of MMD. Among these, TC (P-overall < 0.001, P-nonlinear = 0.037), non-HDL-C (P-overall < 0.001, P-nonlinear = 0.027), and LCI (P-overall = 0.006, P-nonlinear = 0.025) demonstrated significant nonlinear dose-response relationships. By contrast, although LDL-C, AC, CRI-I, and CRI-II showed significant overall associations, their tests for nonlinearity were not statistically significant, suggesting that their relationships with the hemorrhagic phenotype were predominantly approximately linear and positive. Overall, the strength of association with the hemorrhagic phenotype increased with increasing levels of cholesterol-related traditional and nontraditional lipid parameters (Fig. 3 ). Discriminative and Incremental Predictive Value of Traditional and Nontraditional Lipid Parameters for the Hemorrhagic Phenotype of MMD ROC analysis showed that among the traditional lipid parameters, TC and LDL-C had modest discriminative ability for the hemorrhagic phenotype of MMD, with AUCs of 0.637 (95% CI: 0.602–0.671) and 0.633 (95% CI: 0.598–0.669), respectively. Among the nontraditional lipid parameters, non-HDL-C performed best, with an AUC of 0.624 (95% CI: 0.589–0.659), whereas the AUCs of the other markers were generally lower (Fig. 4 ). Based on these results, TC, LDL-C, non-HDL-C, and CRI-II were further added individually to the basic clinical model to evaluate their incremental predictive value. All four markers improved model discrimination, with the largest increase observed for LDL-C, which improved the C statistic from 0.669 to 0.710. At the same time, TC, LDL-C, and non-HDL-C all showed relatively stable improvements in reclassification, with NRIs of 0.394 (95% CI: 0.264–0.523), 0.408 (95% CI: 0.274–0.559), and 0.389 (95% CI: 0.279–0.524), and corresponding IDIs of 0.0312 (95% CI: 0.0140–0.0507), 0.0349 (95% CI: 0.0146–0.0581), and 0.0294 (95% CI: 0.0164–0.0525), respectively (Table 2 ). DeLong testing showed that all four lipid-enhanced models were significantly superior to the basic model. Further pairwise comparisons indicated that the LDL-C-enhanced and non-HDL-C-enhanced models both had significantly better discrimination than the CRI-II-enhanced model, whereas the TC-enhanced model did not differ significantly from either the LDL-C-enhanced or the non-HDL-C-enhanced model. Overall, LDL-C and non-HDL-C showed relatively better incremental predictive performance, whereas the overall discriminative performance of TC, LDL-C, and non-HDL-C was similar (Supplementary Table 2). Table 2 Incremental predictive value of lipid parameters added to the basic clinical model Variables C statistics Estimate (95% CI) P value NRI Estimate (95% CI) P value IDI Estimate (95% CI) P value Basic model 0.669 (0.635–0.704) < 0.001 Reference Reference Reference Reference Basic model + TC 0.706 (0.673–0.739) < 0.001 0.394 (0.264–0.523) < 0.001 0.0312 (0.0140–0.0507) 0.002 Basic model + LDL-C 0.710 (0.677–0.743) < 0.001 0.408 (0.274–0.559) < 0.001 0.0349 (0.0146–0.0581) 0.002 Basic model + Non-HDL-C 0.706 (0.674–0.739) < 0.001 0.389 (0.279–0.524) < 0.001 0.0294 (0.0164–0.0525) 0.004 Basic model + CRI-II 0.694 (0.660–0.727) < 0.001 0.219 (0.085–0.334) 0.001 0.0185 (0.0073–0.0337) 0.027 Basic model + AC 0.687 (0.654–0.721) < 0.001 0.180 (0.085–0.351) 0.008 0.0138 (0.0032–0.0310) 0.052 Basic model + CRI-I 0.687 (0.654–0.721) < 0.001 0.180 (0.051–0.344) 0.016 0.0138 (0.0024–0.0276) 0.032 Basic model + LCI 0.677 (0.642–0.711) < 0.001 0.197 (0.034–0.321) 0.012 0.0042 (0.0000-0.0197) 0.413 Note: The basic model included sex, age, BMI, hypertension, diabetes, smoking status, drinking status, bilateral involvement, and dichotomized Suzuki stage ( 4). Missing BMI values were handled by multiple imputation (5 imputations) before model fitting. C statistics were reported as AUCs with 95% confidence intervals based on the nonparametric DeLong method using pooled predicted probabilities. Continuous NRI and IDI 95% confidence intervals and P values were obtained by bootstrap resampling (100 resamples). Sensitivity and Subgroup Analyses Sensitivity analyses showed that, after repeating the analyses in the complete-case dataset, the main findings remained largely consistent with those of the primary analysis after multiple imputation. TC, LDL-C, non-HDL-C, AC, CRI-I, CRI-II, and LCI remained significantly associated with hemorrhagic MMD, whereas TG, HDL-C, AIP, RC, and RC/HDL-C still showed no stable associations (Supplementary Table 3). After further subdividing the non-hemorrhagic group into asymptomatic and ischemic subgroups, the comparison between the hemorrhagic and ischemic groups was highly consistent with the main analysis (Supplementary Table 5), whereas the comparison with the asymptomatic group still showed significant associations for TC, LDL-C, and non-HDL-C, although the effects of some markers were attenuated (Supplementary Table 4). Subgroup analyses further showed that TC, LDL-C, and non-HDL-C remained positively associated with hemorrhagic MMD across strata of age, sex, hypertension status, and BMI, and none of the interaction tests reached statistical significance, supporting the robustness of these associations (Fig. 5 ; Supplementary Table 6). Mediation Analysis To further investigate the potential mediating role of BMI in the associations between lipid parameters and the hemorrhagic phenotype of MMD, mediation analyses were performed separately for each traditional and nontraditional lipid parameter. The analyses showed that BMI exerted significant mediation effects for some lipid parameters, predominantly in a negative direction. Specifically, significant negative indirect effects were observed for non-HDL-C, AC, CRI-I, and CRI-II. For non-HDL-C, the indirect effect was beta = -0.0370 (P = 0.014), with a proportion mediated of -9.40% (95% CI: -17.7% to -1.1%). For AC and CRI-I, the indirect effects were both beta = -0.0644 (P < 0.001), and the proportions mediated were both − 30.77% (95% CI: -55.8% to -5.7%). For CRI-II, the indirect effect was beta = -0.0799 (P = 0.001), with a proportion mediated of -26.14% (95% CI: -46.8% to -5.5%) (Fig. 6 ; Supplementary Table 7). By contrast, no significant mediation effects were observed for the remaining lipid parameters. These findings suggest that BMI may act more as a suppressive mediator than a conventional concordant mediator in the associations between key nontraditional lipid parameters and the hemorrhagic phenotype of MMD. Discussion This dual-center cross-sectional study of adults with primary MMD systematically evaluated the associations between traditional and nontraditional lipid parameters and the hemorrhagic phenotype, and further examined their dose-response patterns, discriminative and incremental predictive value, and the potential mediating role of BMI. The main findings were as follows. First, after full adjustment for confounding factors, TC, LDL-C, non-HDL-C, AC, CRI-I, CRI-II, and LCI remained significantly associated with the hemorrhagic phenotype, with TC, LDL-C, and non-HDL-C showing the most stable associations and the largest effect sizes. Second, RCS analysis suggested nonlinear dose-response associations of TC, non-HDL-C, and LCI with the hemorrhagic phenotype, and the addition of TC, LDL-C, and non-HDL-C to the basic clinical model improved both model discrimination and reclassification. Third, BMI showed a significant negative mediating effect in the associations between several key lipid parameters and the hemorrhagic phenotype. Overall, these findings suggest that the hemorrhagic phenotype of adult primary MMD may correspond to a lipid metabolic pattern characterized by cholesterol-related lipoprotein abnormalities and may provide supplementary information for risk stratification. One of the most important findings of the present study is the relatively consistent and stable association between cholesterol-related lipid markers and the hemorrhagic phenotype of MMD. Compared with TG, RC, RC/HDL-C, and AIP, TC, LDL-C, and non-HDL-C showed stronger and more stable statistical signals across baseline comparisons, multivariable logistic regression, RCS analysis, and incremental predictive analysis. In the fully adjusted model, the ORs for the highest versus the lowest quartiles of TC, LDL-C, and non-HDL-C were all close to threefold, indicating a fairly substantial association with the hemorrhagic phenotype. Although composite indices such as AC, CRI-I, CRI-II, and LCI also remained significant, they did not clearly outperform these three core markers. Taken together, in the special cerebrovascular context of MMD, the lipid abnormalities that are more stably linked to the hemorrhagic phenotype may primarily reflect an increased burden of cholesterol-related lipoproteins and imbalance in lipoprotein composition, rather than isolated hypertriglyceridemia or remnant cholesterol abnormalities. This finding should be interpreted in light of the disease-specific pathophysiological background of MMD. In adult MMD, particularly hemorrhagic MMD, the core pathology is not classic large-artery atherosclerosis, but rather progressive stenosis or occlusion of the terminal internal carotid arteries and proximal branches, accompanied by the development of abnormal collateral vascular networks, hemodynamic redistribution, and dysregulated vascular wall remodeling[ 3 , 18 ]. Previous AHA/ASA scientific statements and related reviews have pointed out that hemorrhagic MMD in adults is closely associated with fragile collateral circulation and vulnerable periventricular anastomoses, which are regarded as important structural bases for hemorrhage[ 3 , 18 ]. More recent studies have further shown that periventricular anastomoses are not only an important source of hemorrhage but are also closely related to the risk of rebleeding[ 18 , 19 ]. Therefore, in this disease setting, metabolic factors that affect vascular wall stability and fragility may have clinical implications distinct from those in general atherosclerotic disease. Against this background, the increased burden of cholesterol-related lipid parameters observed in the present study may not simply represent conventional atherosclerotic risk factors, but may instead indicate a metabolic milieu that promotes vascular wall injury. Previous studies have shown that the apoB-containing atherogenic lipoprotein particles represented by LDL-C and non-HDL-C more directly reflect the burden of injurious lipid exposure to the vascular wall than TG-related markers. Under oxidative and inflammatory conditions, these lipoproteins may reduce nitric oxide bioavailability, increase reactive oxygen species generation, and upregulate inflammatory and adhesion molecules, thereby impairing endothelial homeostasis and weakening vascular wall integrity[ 20 , 21 , 22 ]. In patients with MMD, who already have abnormal collateral formation, increased hemodynamic burden, and vascular wall fragility[ 3 , 23 ], such a cholesterol-related vascular injury background may further aggravate local vascular instability. In addition, oxidative stress may affect cerebral microvascular endothelial tight junctions and blood-brain barrier stability, thereby increasing the vulnerability of fragile collateral vessels[ 24 , 25 ]. Although these mechanisms have not yet been directly confirmed in MMD[ 2 ], general biological mechanisms of cerebrovascular injury suggest that they may represent an important link between cholesterol-related lipid burden and the hemorrhagic phenotype[ 24 ]. This interpretation is also consistent with the direction of recent studies on lipid metabolism in MMD. One study reported a broader spectrum of lipid abnormalities in MMD, including elevated oxLDL, sdLDL, and Lp(a), with oxLDL showing a relatively strong independent risk signal, suggesting that lipid metabolic abnormalities may participate in the development and progression of MMD[ 5 ]. The present findings also suggest that not all lipid parameters carry equal value for the hemorrhagic phenotype of MMD. TG, RC, RC/HDL-C, and AIP did not show stable associations after full adjustment, whereas TC, LDL-C, and non-HDL-C remained consistently significant. This difference suggests that the metabolic background corresponding to the hemorrhagic phenotype of adult MMD may differ from the classic metabolic syndrome or insulin-resistance phenotype dominated by hypertriglyceridemia[ 26 , 27 ], and may instead be more closely related to the vascular injury pattern reflected by cholesterol-related lipoprotein abnormalities. At the same time, the observation that AIP was lower in the hemorrhagic group at baseline is also noteworthy. AIP primarily reflects the relationship between TG and HDL-C[ 28 ]; in the present study, the hemorrhagic group did not have substantially higher TG but did have relatively higher HDL-C, making its direction different from those of TC, LDL-C, and non-HDL-C statistically understandable. In addition, higher HDL-C concentrations do not necessarily indicate a parallel enhancement of anti-inflammatory, antioxidant, and endothelial-protective functions, because HDL may become dysfunctional under inflammatory or oxidative stress conditions[ 29 , 30 ]. RCS analysis and incremental predictive analysis further support the notion that TC, LDL-C, and non-HDL-C are more informative than other lipid parameters for characterizing the hemorrhagic phenotype of MMD. RCS analysis showed significant nonlinear associations of TC, non-HDL-C, and LCI with the hemorrhagic phenotype, whereas LDL-C, AC, CRI-I, and CRI-II mainly showed approximately linear positive associations, suggesting that different lipid parameters may have threshold effects or accelerated risk ranges in relation to the hemorrhagic phenotype. At the same time, ROC and incremental analyses showed that although the individual discriminative ability of TC, LDL-C, and non-HDL-C was only moderate, the model AUC increased from 0.669 to 0.706–0.710 after the inclusion of these markers in the basic clinical model, with statistically significant improvements in both NRI and IDI; among them, LDL-C showed the largest incremental gain. For MMD, a disease that lacks mature biomarkers[ 31 ], readily available lipid measures obtained at admission, with low cost and high reproducibility, hold practical translational potential. Sensitivity and subgroup analyses further support the stability of the observed associations. The consistency between complete-case analyses and analyses after multiple imputation suggests that the main conclusions do not depend on the handling of missing BMI values. After the non-hemorrhagic group was further divided into asymptomatic and ischemic subgroups, the results comparing the hemorrhagic and ischemic groups were highly consistent with those of the main analysis, whereas some marker effects were attenuated in comparisons with the asymptomatic group. This pattern suggests that the lipid differences identified in the present study may more likely reflect biological divergence between hemorrhagic and ischemic phenotypes. In addition, TC, LDL-C, and non-HDL-C remained directionally consistent across strata of age, sex, hypertension, and BMI, and no significant interactions were observed, indicating that these associations were robust across clinical subgroups. The present study also found that BMI had a negative mediating effect in the associations between several key lipid parameters and the hemorrhagic phenotype. This pattern is inconsistent with the classic pathway in conventional cardiovascular research, in which higher BMI increases event risk by aggravating lipid metabolic disturbances[ 32 ]. Instead, it is more compatible with a suppression effect: lipid parameters were positively associated with BMI, whereas BMI was negatively associated with the hemorrhagic phenotype of MMD, such that the BMI-related pathway statistically attenuated the overall effects of lipid parameters on the hemorrhagic phenotype[ 33 , 34 ]. Several large cohort studies have also reported that lower BMI is associated with an increased risk of hemorrhagic stroke[ 35 , 36 , 37 ]. In addition, previous studies have shown that individuals with lower BMI are more prone to cerebral microbleeds[ 38 ], which are regarded as imaging markers of cerebral small-vessel fragility and may, to some extent, resemble the hemorrhagic pattern of newly formed vessels in MMD. In the baseline characteristics of the present study, patients in the hemorrhagic group had lower BMI but higher levels of several cholesterol-related lipid parameters. This finding suggests that the meaning of BMI in MMD may differ from that of obesity exposure in the general population and may, to some extent, reflect nutritional reserve, vascular repair capacity, or vascular fragility. From a clinical translation perspective, the present study has at least three potential implications. First, TC, LDL-C, and non-HDL-C are all derived from routine lipid testing and have the advantages of low cost, wide availability, and good reproducibility[ 39 , 40 ], making them potentially useful as supplementary biological information in the admission evaluation of adults with MMD. Second, compared with single lipid parameters, non-HDL-C and several composite markers may better reflect the burden of lipoproteins and imbalance in lipoprotein composition[ 41 , 20 ]. Third, given their incremental value beyond the basic clinical model, these lipid parameters could be integrated in the future with imaging risk markers, hemodynamic parameters, and clinical variables to develop a more comprehensive risk assessment model for hemorrhage in adult MMD. The present study also has several limitations. First, this was a retrospective dual-center cross-sectional observational study. Although multivariable adjustment, sensitivity analyses, and subgroup analyses were performed, residual confounding cannot be completely excluded, and causal inference cannot be established. Second, the present analysis focused on the hemorrhagic phenotype of adult primary MMD rather than incident hemorrhagic events during prospective follow-up. Therefore, the findings are more applicable to phenotypic differentiation and risk stratification than to prediction of future hemorrhage. Third, all lipid parameters were derived from a single measurement at admission, which cannot reflect long-term lipid exposure or dynamic changes over time and cannot fully rule out the influence of acute events or stress states on metabolic markers. Fourth, although major clinical variables and Suzuki stage were included in the adjustment models, more detailed imaging risk markers and some treatment-related factors, such as prior lipid-lowering therapy, antiplatelet therapy, or more detailed revascularization information, were unavailable. Accordingly, the specific pathways linking lipid parameters with the hemorrhagic phenotype still require further investigation. Future multicenter prospective studies, combined with dynamic lipid monitoring, metabolomics or lipidomics, and vascular wall imaging and cerebral blood flow assessment, are still needed to further clarify the role of cholesterol-related lipid parameters in phenotypic differentiation and hemorrhagic risk assessment in MMD. Conclusion In summary, the hemorrhagic phenotype of adult primary MMD was closely associated with elevated cholesterol-related lipid parameters, among which TC, LDL-C, and non-HDL-C showed the most stable associations. The negative mediating effect of BMI further suggests that the relationship between metabolic status and the hemorrhagic phenotype in MMD may differ from that observed in general cardio-cerebrovascular diseases. These findings provide new clues for mechanistic research and risk assessment in the hemorrhagic phenotype of adult primary MMD. Declarations Competing interests The authors declare no conflicts of interest. Ethical approval This study was approved by the Ethics Committee of Beijing Hospital (2024BJYYEC-KY164-04) and the Ethics Committee of Beijing Tiantan Hospital, Capital Medical University (KY-2022-051-08). The study was conducted in accordance with the Declaration of Helsinki. Consent to participate Because of the retrospective nature of the study, the requirement for informed consent was waived by the ethics committees. Author Contribution Zhang Dong, Weihong Huang and Haoyang Xiong conceived and designed the study. Wei Liu, Peicong Ge, Wenjing Chen, Yunhao Liu, Xuefeng Tang, Ziheng Yang, and Zhongji Yan collected the data. Weihong Huang and Wenjing Chen drafted the manuscript. All authors critically revised the manuscript and approved the final version. Acknowledgement We thank the individuals who contributed to the study or manuscript preparation but did not fulfill all the criteria of authorship. Data availability statement The datasets generated and analyzed during the current study are not publicly available because of patient privacy and institutional restrictions, but are available from the corresponding author on reasonable request. References Diagnostic. Criteria for Moyamoya Disease – 2021 Revised Version. Neurol Med Chir (Tokyo). 2022;62(7):307–12. https://doi.org/10.2176/jns-nmc.2022-0072 . Moyamoya Disease: Pathophysiology, Diagnosis, and Treatment. Dtsch Arztebl Int. 2025;122(26):722–8. https://doi.org/10.3238/arztebl.m2025.0185 Adult Moyamoya Disease and Syndrome. Current Perspectives and Future Directions: A Scientific Statement From the American Heart Association/American Stroke Association. Stroke. 2023;54(10):e465–79. https://doi.org/10.1161/STR.0000000000000443 . Hemorrhagic Moyamoya Disease. A Recent Update. J Korean Neurosurg Soc. 2019;62(2):136–43. https://doi.org/10.3340/jkns.2018.0101 . Unraveling the Dyslipidemic Landscape in Moyamoya Disease: OxLDL as a Key Biomarker. CNS Neurosci Ther. 2025;31(5):e70441. https://doi.org/10.1111/cns.70441 Hypo-high density lipoproteinemia is a predictor for recurrent stroke during the long-term follow-up after revascularization in adult moyamoya disease. Front Neurol. 2022;13:891622. https://doi.org/10.3389/fneur.2022.891622 Data-Independent Acquisition-Based Serum Proteomic Profiling of Adult Moyamoya Disease Patients Reveals the Potential Pathogenesis of Vascular Changes. J Mol Neurosci. 2022;72(12):2473–85. https://doi.org/10.1007/s12031-022-02092-w Association Between the Onset Pattern of Adult Moyamoya Disease and Risk Factors for Stroke. Stroke. 2020;51(10):3124–8. https://doi.org/10.1161/STROKEAHA.120.030653 . 2021 Japanese Guidelines for the Management of Moyamoya Disease: Guidelines from the Research Committee on Moyamoya Disease and Japan Stroke Society. Neurol Med Chir (Tokyo). 2022;62(4):165–170. https://doi.org/10.2176/jns-nmc.2021-0382 Nonlinear relationship between untraditional lipid parameters and the risk of prediabetes: a large retrospective study based on Chinese adults. Cardiovasc Diabetol. 2024;23(1):12. https://doi.org/10.1186/s12933-023-02103-z ApoB LDL-C. and non-HDL-C as markers of cardiovascular risk. J Clin Lipidol. 2025;19(4):844–59. https://doi.org/10.1016/j.jacl.2025.05.024 . Potential Association Between Atherogenic Coefficient. Prognostic Nutritional Index, and Various Obesity Indices in Diabetic Nephropathy. Nutrients. 2025;17(8):1339. https://doi.org/10.3390/nu17081339 . Dotinurad restores exacerbated kidney dysfunction in hyperuricemic patients with chronic kidney disease. BMC Nephrol. 2024;25(1):97. https://doi.org/10.1186/s12882-024-03535-9 The lipid ratio castelli's risk index II is a novel biomarker for intraplaque neovascularization in patients with carotid stenosis. Lipids Health Dis. 2025;25(1):11. https://doi.org/10.1186/s12944-025-02821-1 The effects of a novel nutraceutical combination on low-density lipoprotein cholesterol and other markers of cardiometabolic health in adults with hypercholesterolaemia: A randomised double-blind placebo-controlled trial. Atherosclerosis. 2025;403:119177. https://doi.org/10.1016/j.atherosclerosis.2025.119177 Huang X, Li C, Zeng P, Ling Y, Tan S, Bai Z, Shen S, Chen S, Nie B, Wang H, Lyu J. Non-traditional lipid-inflammatory parameters estimate the risk of stroke in middle-aged and older Chinese adults: a nationwide prospective cohort study. J Adv Res 2025 Nov 17:S2090-1232(25)00924–5. https://doi.org/10.1016/j.jare.2025.11.029 Lipid metabolism. BMI and the risk of nonalcoholic fatty liver disease in the general population: evidence from a mediation analysis. J Transl Med. 2023;21(1):192. https://doi.org/10.1186/s12967-023-04047-0 . [Adult Hemorrhagic Moyamoya Disease. Physiopathology and Treatment]. No Shinkei Geka. 2025;53(3):576–82. https://doi.org/10.11477/mf.030126030530030576 . High rebleeding risk associated with choroidal collateral vessels in hemorrhagic moyamoya disease: analysis of a nonsurgical cohort in the Japan Adult Moyamoya Trial. J Neurosurg. 2019;130(2):525–30. https://doi.org/10.3171/2017.9.JNS17576 The Role of Non-HDL Cholesterol and Apolipoprotein B in Cardiovascular Disease: A Comprehensive Review. J Cardiovasc Dev Dis. 2025;12(7):256. https://doi.org/10.3390/jcdd12070256 Unravelling the Mechanisms of Oxidised Low-Density Lipoprotein in Cardiovascular Health: Current Evidence from In Vitro and In Vivo Studies. Int J Mol Sci. 2024;25(24):13292. https://doi.org/10.3390/ijms252413292 MedComm. (2020). 2024;5(8):e651. https://doi.org/10.1002/mco2.651 Temporal Base Transdural Anastomosis in Moyamoya Disease. A Potential Association with Posterior Cerebral Artery Involvement and Clinical Importance. Neurol Med Chir (Tokyo). 2025;65(8):366–72. https://doi.org/10.2176/jns-nmc.2025-0113 . The role of oxidative stress in blood-brain barrier disruption during ischemic stroke: Antioxidants in clinical trials. Biochem Pharmacol. 2024;228:116186. https://doi.org/10.1016/j.bcp.2024.116186 Aging. vascular dysfunction, and the blood-brain barrier: unveiling the pathophysiology of stroke in older adults. Biogerontology. 2025;26(2):67. https://doi.org/10.1007/s10522-025-10209-y . Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. 2009;120(16):1640–1645. https://doi.org/10.1161/CIRCULATIONAHA.109.192644 Adipose Tissue Insulin Resistance. A Key Driver of Metabolic Syndrome Pathogenesis. Biomedicines. 2025;13(10):2376. https://doi.org/10.3390/biomedicines13102376 . Atherogenic. index of plasma [log(triglycerides/HDL-cholesterol)]: theoretical and practical implications. Clin Chem. 2004;50(7):1113–1115. https://doi.org/10.1373/clinchem.2004.033175 HDL in the 21st Century: A Multifunctional Roadmap for Future HDL Research. Circulation. 2021;143(23):2293–2309. https://doi.org/10.1161/CIRCULATIONAHA.120.044221 Dysfunctional high-density lipoprotein: an updated review. Front Cardiovasc Med. 2025;12:1713387. https://doi.org/10.3389/fcvm.2025.1713387 Molecular. and multimodal biomarkers in Moyamoya disease: from pathogenic mechanisms to clinical translation. Eur J Med Res. 2026;31(1):187. https://doi.org/10.1186/s40001-025-03769-9 Obesity and Cardiovascular Disease. A Scientific Statement From the American Heart Association. Circulation. 2021;143(21):e984–1010. https://doi.org/10.1161/CIR.0000000000000973 . Equivalence of the mediation, confounding and suppression effect. Prev Sci. 2000;1(4):173–81. https://doi.org/10.1023/a:1026595011371 Reasons for Testing Mediation in the Absence of an Intervention Effect. A Research Imperative in Prevention and Intervention Research. J Stud Alcohol Drugs. 2018;79(2):171–81. https://doi.org/10.15288/jsad.2018.79.171 . Adiposity. risk of ischaemic and haemorrhagic stroke in 0.5 million Chinese men and women: a prospective cohort study. Lancet Glob Health. 2018;6(6):e630–40. https://doi.org/10.1016/S2214-109X(18)30216-X . Adiposity. ischemic and hemorrhagic stroke: Prospective study in women and meta-analysis. Neurology. 2016;87(14):1473–81. https://doi.org/10.1212/WNL.0000000000003171 . Clinical impact of body mass index on outcomes of ischemic and hemorrhagic strokes. Int J Stroke. 2024;19(8):907–15. https://doi.org/10.1177/17474930241249370 Severe underweight and cerebral microbleeds. J Neurol. 2012;259(12):2707–13. https://doi.org/10.1007/s00415-012-6574-7 2026 ACC/AHA/AACVPR/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Dyslipidemia: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2026 Mar 13. https://doi.org/10.1161/CIR.0000000000001423 [Chinese guidelines for lipid management. (2023)]. Zhonghua Xin Xue Guan Bing Za Zhi. 2023;51(3):221–255. https://doi.org/10.3760/cma.j.cn112148-20230119-00038 Residual Atherosclerotic Cardiovascular Disease Risk. Focus on Non-High-Density Lipoprotein Cholesterol. J Cardiovasc Pharmacol Ther. 2023;28:10742484231189597. https://doi.org/10.1177/10742484231189597 . Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterials.pdf Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 12 May, 2026 Reviews received at journal 10 May, 2026 Reviewers agreed at journal 29 Apr, 2026 Reviewers agreed at journal 28 Apr, 2026 Reviewers agreed at journal 28 Apr, 2026 Reviewers invited by journal 27 Apr, 2026 Editor assigned by journal 18 Apr, 2026 Submission checks completed at journal 13 Apr, 2026 First submitted to journal 12 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9393575","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":631737656,"identity":"0045429d-ce06-494b-9924-6da99299cc63","order_by":0,"name":"Weihong Huang","email":"","orcid":"","institution":"Chinese Academy of Medical Sciences \u0026 Peking Union Medical College","correspondingAuthor":false,"prefix":"","firstName":"Weihong","middleName":"","lastName":"Huang","suffix":""},{"id":631737657,"identity":"f5958985-00c8-40f9-9623-25b6d847e738","order_by":1,"name":"Haoyang Xiong","email":"","orcid":"","institution":"Chinese Academy of Medical Sciences \u0026 Peking Union Medical College","correspondingAuthor":false,"prefix":"","firstName":"Haoyang","middleName":"","lastName":"Xiong","suffix":""},{"id":631737658,"identity":"f227adb3-bd8a-49c3-8fbd-50b6719dedc8","order_by":2,"name":"Wenjing Chen","email":"","orcid":"","institution":"Southwest Medical University","correspondingAuthor":false,"prefix":"","firstName":"Wenjing","middleName":"","lastName":"Chen","suffix":""},{"id":631737659,"identity":"59e43100-b6f6-4a96-8333-efc90e495205","order_by":3,"name":"Wei Liu","email":"","orcid":"","institution":"Beijing Tiantan Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Liu","suffix":""},{"id":631737660,"identity":"88e2cb12-1a5f-40c4-a6a5-5145e1bb4585","order_by":4,"name":"Peicong Ge","email":"","orcid":"","institution":"Beijing Tiantan Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Peicong","middleName":"","lastName":"Ge","suffix":""},{"id":631737661,"identity":"457b85be-c285-4d62-9b9c-c6223a3e44fd","order_by":5,"name":"Yunhao Liu","email":"","orcid":"","institution":"Southwest Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yunhao","middleName":"","lastName":"Liu","suffix":""},{"id":631737662,"identity":"a3536bda-d377-46e6-8fa9-75f3065c74d3","order_by":6,"name":"Xuefeng Tang","email":"","orcid":"","institution":"Southwest Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xuefeng","middleName":"","lastName":"Tang","suffix":""},{"id":631737663,"identity":"eabbcb9d-7fa9-4d6e-b455-77cd23f6614f","order_by":7,"name":"Ziheng Yang","email":"","orcid":"","institution":"Southwest Medical University","correspondingAuthor":false,"prefix":"","firstName":"Ziheng","middleName":"","lastName":"Yang","suffix":""},{"id":631737664,"identity":"aff691dc-d6f8-49a9-9f09-36b6c906d700","order_by":8,"name":"Zhongji Yan","email":"","orcid":"","institution":"Southwest Medical University","correspondingAuthor":false,"prefix":"","firstName":"Zhongji","middleName":"","lastName":"Yan","suffix":""},{"id":631737665,"identity":"ddb5dfb9-7a95-4e05-9c62-981abadba022","order_by":9,"name":"Dong Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBklEQVRIiWNgGAWjYDACCRDBBsTMzAcfJFQwyLARr4W9LdngwxkGHhK08JxRE5zZxsBDUIf87OZnD7+U2eTJR+SwMfPOs+Phk8gxe8BQYxONSwvjnGPmxjLn0ooNb+Qee8y7LZmHTSLH3IDhWFpuAw4tzBIJZtKSbYcTN87ISzfm3cYM0mImwdhwGKcWNon0b1AtOWbSvHPqCWvhASqQ/AjUMp/njJnkzIbDhLVISOSUSTOcS0vcAA7kY8d52HielUkk4PGL/Iz0bZI/ymwS5zeDorKmWk6+PXmbxIcaG5xawEEAiguDAzCuQAIDQwIe5SDA+ANkHdxQ/gM4VY6CUTAKRsHIBACjbVU6iSvUEgAAAABJRU5ErkJggg==","orcid":"","institution":"Chinese Academy of Medical Sciences \u0026 Peking Union Medical College","correspondingAuthor":true,"prefix":"","firstName":"Dong","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2026-04-12 10:53:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9393575/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9393575/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108632444,"identity":"36fdddef-7512-4822-a930-40bac6280e0b","added_by":"auto","created_at":"2026-05-06 17:05:12","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":371416,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFlowchart of study participants.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9393575/v1/d732d3205647e97584081136.jpeg"},{"id":108806006,"identity":"8c29f146-2818-48ff-b6cd-b7717150b360","added_by":"auto","created_at":"2026-05-08 15:27:26","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":167568,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAssociations of lipid parameters with the risk of hemorrhage.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNote: Models were adjusted for sex, age, BMI, hypertension, diabetes, smoking status, drinking status, bilateral involvement, and Suzuki stage. Abbreviations: OR, odds ratio; CI, confidence interval; AIP, atherogenic index of plasma; LCI, lipoprotein combined index; CRI-I, Castelli's index-I; CRI-II, Castelli's index-II; RC, remnant cholesterol; RC/HDL-C, remnant cholesterol/high-density lipoprotein cholesterol; AC, atherogenic coefficient; Non-HDL-C, non-high-density lipoprotein cholesterol.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9393575/v1/540899d84f4258ba84f3f9b0.png"},{"id":108805653,"identity":"5c26cbfb-2f64-4f86-8eef-0d60cbe66716","added_by":"auto","created_at":"2026-05-08 15:26:31","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":144381,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDose-response relationship between the key lipid parameters and the risk of hemorrhage. \u003c/strong\u003eNote: Models were adjusted for sex, age, BMI, hypertension, diabetes, smoking status, drinking status, bilateral involvement, and Suzuki stage.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-9393575/v1/7c649f4fa167ad837dd2ab0a.png"},{"id":108805749,"identity":"bb7c74f1-0c66-4a84-9b7e-c614f91eb188","added_by":"auto","created_at":"2026-05-08 15:26:48","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":168079,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eROC curve analysis of traditional (A) and nontraditional (B) lipid parameters in predicting the risk of hemorrhage.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNote: AUC, area under the curve; CI, confidence interval.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-9393575/v1/c4930e6090555d0dc36e8714.png"},{"id":108804981,"identity":"87d417c2-0217-45e4-9c9f-73552168e921","added_by":"auto","created_at":"2026-05-08 15:24:26","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":180122,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSubgroup analysis of the associations between lipid parameters and the risk of hemorrhage stratified by age, sex, hypertension, and BMI.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNote: Models were adjusted for sex, age, BMI, hypertension, diabetes, smoking status, drinking status, bilateral involvement, and Suzuki stage.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-9393575/v1/a5840ed461cee237e41259c7.png"},{"id":108804958,"identity":"8f9b56ef-0e1f-4ad0-9633-07b48d0cb3c5","added_by":"auto","created_at":"2026-05-08 15:24:21","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":126161,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMediation analysis of BMI in the associations between selected lipid parameters and the risk of hemorrhage.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNote: Models were adjusted for age, sex, hypertension, diabetes, smoking, alcohol consumption, laterality, and Suzuki stage. *P \u0026lt; 0.05, **P \u0026lt; 0.01, ***P \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-9393575/v1/fceb4620d6bb7fe243822157.png"},{"id":108812196,"identity":"042f1185-8cb2-4f46-bc4a-16f2fea9f41c","added_by":"auto","created_at":"2026-05-08 16:09:47","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1365248,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9393575/v1/573ac2ab-3158-466d-81e6-f31d29712f48.pdf"},{"id":108632442,"identity":"709961ea-e344-4060-ba4b-c41a99b8a772","added_by":"auto","created_at":"2026-05-06 17:05:12","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1440192,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterials.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9393575/v1/8178b74006999e2d08b76c20.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Association of Traditional and Novel Lipid Indicators With the Hemorrhagic Phenotype in Adult Moyamoya Disease: Implications for Lipid Risk Stratification","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMoyamoya disease (MMD) is a chronic cerebrovascular disorder characterized by progressive stenosis or occlusion of the terminal internal carotid arteries and the proximal anterior and middle cerebral arteries, accompanied by the formation of an abnormal collateral vascular network at the base of the brain.[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] In adults, stroke is the major clinical outcome, and the hemorrhagic phenotype is associated with higher disability and mortality. Its occurrence is closely related to fragile collateral circulation, abnormal hemodynamic burden, and dysregulated vascular wall remodeling[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. In recent years, in addition to genetic susceptibility and perfusion abnormalities, lipid metabolic disturbances have also been suggested to participate in the development and progression of MMD[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Previous studies have shown that patients with MMD may exhibit abnormalities not only in traditional lipid indices but also in nontraditional lipid markers such as oxLDL, sdLDL, Lp(a), and non-HDL-C, some of which are associated with disease progression or the risk of cerebrovascular events[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. These findings suggest that lipid abnormalities may contribute to the pathophysiology of MMD through oxidative stress, endothelial injury, inflammatory activation, and abnormal vascular remodeling[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eHowever, existing evidence has mainly focused on the overall risk of MMD[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], postoperative recurrent stroke[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e], or single metabolic factors[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Systematic studies examining the lipid metabolic profile reflected jointly by traditional and nontraditional lipid parameters in relation to hemorrhagic risk in adults with MMD, particularly the hemorrhagic phenotype, remain limited. Therefore, the present study jointly analyzed traditional and nontraditional lipid parameters to identify lipid metabolic features associated with the hemorrhagic phenotype in adults with MMD and to explore their potential value in hemorrhagic risk assessment and prediction.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy Population\u003c/h2\u003e \u003cp\u003eThis was a retrospective dual-center cross-sectional observational study. We consecutively screened 1,901 patients with MMD who were treated at Beijing Hospital and Beijing Tiantan Hospital, Capital Medical University, between January 2022 and January 2026, including 665 patients from Beijing Hospital and 1,236 from Beijing Tiantan Hospital. All patients were diagnosed with MMD based on clinical manifestations and cerebrovascular imaging findings. Moyamoya syndrome was excluded according to the criteria of the Research Committee on Spontaneous Occlusion of the Circle of Willis and the 2021 Japanese Guidelines for the Management of Moyamoya Disease[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], and therefore all included cases represented primary MMD.\u003c/p\u003e \u003cp\u003eThe inclusion criteria were as follows: (1) diagnosis of MMD based on digital subtraction angiography (DSA); (2) complete demographic, clinical, and imaging data; and (3) lipid testing performed within 24 h of admission, with complete measurements of total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C). The exclusion criteria were: (1) missing key lipid measurements (TC, TG, HDL-C, or LDL-C; n\u0026thinsp;=\u0026thinsp;470); (2) age\u0026thinsp;\u0026lt;\u0026thinsp;18 years (n\u0026thinsp;=\u0026thinsp;243); and (3) incomplete clinical or imaging data that precluded subsequent statistical analysis (n\u0026thinsp;=\u0026thinsp;12). Ultimately, 1,176 adults with MMD were included in the final analysis, comprising 857 patients with a non-hemorrhagic phenotype and 319 with a hemorrhagic phenotype (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Hemorrhagic MMD was defined as intracranial hemorrhage confirmed by cranial CT or MRI either at admission or previously, including intracerebral hemorrhage, intraventricular hemorrhage, or subarachnoid hemorrhage. Non-hemorrhagic MMD included asymptomatic cases and ischemic cases, including transient ischemic attack (TIA) and cerebral infarction.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eData Collection\u003c/h3\u003e\n\u003cp\u003eDemographic, clinical, and laboratory data were collected for all included patients. Demographic and clinical variables included age, sex, body mass index (BMI), smoking history, drinking history, hypertension, diabetes mellitus, lesion laterality (unilateral or bilateral), and Suzuki stage. Patients were classified as having a hemorrhagic or non-hemorrhagic phenotype according to clinical presentation at admission and imaging findings. Laboratory data were obtained from routine hematologic and lipid tests performed at admission, including neutrophil count, lymphocyte count, monocyte count, platelet count, TG, TC, LDL-C, and HDL-C. Imaging data were independently reviewed in a blinded manner by two experienced neurosurgeons; when discrepancies occurred, a third physician adjudicated the final result.\u003c/p\u003e\n\u003ch3\u003eExposure Definitions\u003c/h3\u003e\n\u003cp\u003eThe exposures of interest consisted of baseline traditional lipid parameters and their derived nontraditional lipid parameters. Traditional lipid parameters included TC, TG, HDL-C, and LDL-C, all obtained from the first laboratory test performed after admission. Based on these measurements, the following nontraditional lipid parameters were calculated for subsequent analyses: atherogenic index of plasma (AIP), non-HDL-C, atherogenic coefficient (AC), Castelli's index-I (CRI-I), Castelli's index-II (CRI-II), lipoprotein combined index (LCI), remnant cholesterol (RC), and the RC/HDL-C ratio. The formulas were as follows:\u003c/p\u003e \u003cp\u003e(1) AIP\u0026thinsp;=\u0026thinsp;lg(TG / HDL-C) [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/p\u003e \u003cp\u003e(2) non-HDL-C\u0026thinsp;=\u0026thinsp;TC - HDL-C [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/p\u003e \u003cp\u003e(3) AC\u0026thinsp;=\u0026thinsp;non-HDL-C / HDL-C [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/p\u003e\n\u003cp\u003e(4) CRI-I = TC / HDL-C [13]\u003c/p\u003e\n\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003cp\u003e(5) CRI-II\u0026thinsp;=\u0026thinsp;LDL-C / HDL-C [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/p\u003e \u003cp\u003e(6) LCI\u0026thinsp;=\u0026thinsp;TC x TG x LDL-C / HDL-C [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003cp\u003e(7) RC\u0026thinsp;=\u0026thinsp;TC - LDL-C - HDL-C [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/p\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003cp\u003e(8) RC/HDL-C\u0026thinsp;=\u0026thinsp;RC / HDL-C [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eContinuous variables were first tested for normality using the Shapiro-Wilk test. Because all continuous variables were nonnormally distributed, they are presented as medians with interquartile ranges (IQRs), and between-group comparisons were performed using the Mann-Whitney U test. Categorical variables are presented as counts and percentages and were compared using the chi-square test or Fisher's exact test, as appropriate. Given the lack of established lipid cutoffs for patients with MMD, each lipid parameter was further categorized into quartiles for comparative analyses.\u003c/p\u003e \u003cp\u003eBMI was the only variable with missing data, with 147 missing values among 1,176 patients (12.5%). Missing BMI values were handled using multiple imputation with the mice package, generating five imputed datasets. The imputation model included the outcome variable, lipid parameters, and relevant covariates. Logistic regression models were used to analyze the associations of traditional and nontraditional lipid parameters with hemorrhagic MMD, and odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. Three progressively adjusted models were constructed: Model 1 was unadjusted; Model 2 was adjusted for age, sex, and BMI; and Model 3 was additionally adjusted for hypertension, diabetes, smoking, drinking, bilateral involvement, and Suzuki stage. In quartile analyses, the lowest quartile (Q1) served as the reference group.\u003c/p\u003e \u003cp\u003eTo assess multicollinearity, variance inflation factors (VIFs) were calculated for independent variables in each model, and correlation matrices were generated to evaluate correlations among lipid parameters. A VIF\u0026thinsp;\u0026lt;\u0026thinsp;5 was considered indicative of no significant multicollinearity (Supplementary Fig.\u0026nbsp;1). Restricted cubic spline (RCS) models were further fitted to assess dose-response relationships and potential nonlinear associations between key lipid parameters and hemorrhagic MMD, using four prespecified percentiles as knots and the median value as the reference. RCS analyses were performed with the rms package.\u003c/p\u003e \u003cp\u003eTo evaluate the robustness of the findings, sensitivity analyses were conducted, including repeated analyses in the complete-case dataset and repeated analyses after further dividing the non-hemorrhagic group into asymptomatic and ischemic subgroups (TIA and cerebral infarction). Additional subgroup analyses were performed according to age (stratified at the median age of 42 years), sex, hypertension status, and BMI (\u0026lt;\u0026thinsp;28 vs\u0026thinsp;\u0026gt;\u0026thinsp;=\u0026thinsp;28), and potential effect modification was tested by including multiplicative interaction terms in the main models.\u003c/p\u003e \u003cp\u003eReceiver operating characteristic (ROC) curves and areas under the curve (AUCs) were used to evaluate the discriminative ability of traditional and nontraditional lipid parameters for hemorrhagic MMD, with internal validation by 100 bootstrap resamples. AUCs were compared pairwise using the DeLong test. The incremental predictive value of traditional and nontraditional lipid parameters was further assessed using the integrated discrimination improvement (IDI) and net reclassification improvement (NRI). ROC-related analyses were performed using the pROC package.\u003c/p\u003e \u003cp\u003eIn addition, mediation analysis was conducted using the Rmediation package to evaluate the potential mediating role of BMI in the associations between lipid parameters and hemorrhagic MMD[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. After adjustment for covariates, indirect effects, direct effects, total effects, and the proportion mediated were estimated. All statistical tests were two-sided, and P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant. All analyses were performed using R 4.5.2.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eClinical Characteristics of the MMD Cohort\u003c/h2\u003e \u003cp\u003eA total of 1,176 adults with primary MMD were included, including 857 patients with a non-hemorrhagic phenotype and 319 with a hemorrhagic phenotype. Among the non-hemorrhagic cases, 107 were asymptomatic and 750 were ischemic (including TIA and cerebral infarction). Compared with the non-hemorrhagic group, patients with the hemorrhagic phenotype had lower BMI and lower proportions of hypertension, diabetes, smoking, and bilateral involvement, and they also differed significantly in the distribution of Suzuki stage. By contrast, age, sex, and drinking status did not differ significantly between groups. In terms of lipid parameters, patients with the hemorrhagic phenotype had significantly higher levels of TC, LDL-C, HDL-C, non-HDL-C, AC, CRI-I, CRI-II, and LCI, whereas AIP was significantly lower. No significant between-group differences were observed for TG, RC, or RC/HDL-C (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBaseline characteristics of patients with non-hemorrhagic and hemorrhagic phenotypes\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOverall\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNon-hemorrhagic phenotype\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHemorrhagic phenotype\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eP value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003eDemographic and clinical characteristics\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex, male, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e522 (44.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e391 (45.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e131 (41.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.162\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge, years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42.00 (34.00, 50.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e43.00 (34.00, 51.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e42.00 (34.00, 50.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.625\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI, kg/m\u0026sup2;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e25.00 (22.84, 27.68)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e25.38 (23.15, 28.03)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e24.22 (21.86, 26.38)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypertension, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e448 (38.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e361 (42.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e87 (27.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e161 (13.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e144 (16.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17 (5.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmoking, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e216 (18.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e170 (19.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e46 (14.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.033\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDrinking, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e150 (12.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e119 (13.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e31 (9.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.057\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBilateral involvement, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1035 (88.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e766 (89.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e269 (84.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.018\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSuzuki stage, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026le;\u0026thinsp;4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e888 (75.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e629 (73.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e259 (81.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e288 (24.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e228 (26.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e60 (18.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLipid parameters\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTG, mmol/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.23 (0.85, 1.71)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.25 (0.89, 1.74)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.18 (0.81, 1.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.064\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTC, mmol/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.97 (3.32, 4.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.83 (3.23, 4.50)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.25 (3.68, 4.91)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLDL-C, mmol/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.24 (1.68, 2.82)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.14 (1.58, 2.65)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.49 (1.96, 3.16)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHDL-C, mmol/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.18 (1.00, 1.38)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.16 (0.98, 1.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.23 (1.04, 1.42)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.01 (-0.18, 0.21)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.02 (-0.15, 0.22)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.03 (-0.23, 0.19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNon-HDL-C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.74 (2.14, 3.38)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.64 (2.05, 3.26)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.99 (2.40, 3.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAtherogenic coefficient\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.26 (1.72, 3.02)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.23 (1.68, 2.95)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.38 (1.87, 3.20)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCRI-I\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.26 (2.72, 4.02)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.23 (2.68, 3.95)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.38 (2.87, 4.20)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCRI-II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.85 (1.39, 2.46)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.82 (1.35, 2.39)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.99 (1.52, 2.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLCI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8.87 (4.63, 17.51)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8.61 (4.46, 16.28)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9.69 (5.27, 20.24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.017\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.42 (0.30, 0.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.42 (0.30, 0.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.43 (0.30, 0.58)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.491\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRC/HDL-C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.36 (0.24, 0.53)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.36 (0.24, 0.53)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.37 (0.23, 0.51)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.564\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eNote: BMI, body mass index; TG, triglycerides; TC, total cholesterol; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; LCI, lipoprotein combined index; AIP, atherogenic index of plasma; AC, atherogenic coefficient; CRI-I, Castelli's index-I; CRI-II, Castelli's index-II; RC, remnant cholesterol.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eAssociations of Traditional and Nontraditional Lipid Parameters with Hemorrhagic Risk in MMD\u003c/h2\u003e \u003cp\u003eAfter categorization of traditional and nontraditional lipid parameters into quartiles, logistic regression analyses were performed using the lowest quartile (Q1) as the reference to evaluate the association between the highest quartile (Q4) and hemorrhagic MMD. Multivariable logistic regression showed that, after progressive adjustment for age, sex, BMI, hypertension, diabetes, smoking, drinking, bilateral involvement, and Suzuki stage, TC and LDL-C among traditional lipid parameters, as well as non-HDL-C, AC, CRI-I, CRI-II, and LCI among nontraditional lipid parameters, were all significantly associated with hemorrhagic MMD (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; Supplementary Table\u0026nbsp;1). Among them, TC, LDL-C, and non-HDL-C showed the strongest associations: the ORs for Q4 versus Q1 were 3.435 (95% CI: 2.254\u0026ndash;5.237), 3.197 (95% CI: 2.134\u0026ndash;4.791), and 3.150 (95% CI: 2.087\u0026ndash;4.754), respectively, with all P for trend values\u0026thinsp;\u0026lt;\u0026thinsp;0.001. In contrast, TG and HDL-C among traditional parameters and AIP, RC, and RC/HDL-C among nontraditional parameters were not significantly associated with hemorrhagic MMD after full adjustment. Overall, higher quartile levels of cholesterol-related traditional and nontraditional lipid parameters were closely associated with the hemorrhagic phenotype of MMD, and the strength of association increased across quartiles.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eRestricted Cubic Spline Analysis\u003c/h2\u003e \u003cp\u003eMultivariable-adjusted RCS analyses further showed significant overall associations of TC, LDL-C, non-HDL-C, AC, CRI-I, CRI-II, and LCI with the hemorrhagic phenotype of MMD. Among these, TC (P-overall\u0026thinsp;\u0026lt;\u0026thinsp;0.001, P-nonlinear\u0026thinsp;=\u0026thinsp;0.037), non-HDL-C (P-overall\u0026thinsp;\u0026lt;\u0026thinsp;0.001, P-nonlinear\u0026thinsp;=\u0026thinsp;0.027), and LCI (P-overall\u0026thinsp;=\u0026thinsp;0.006, P-nonlinear\u0026thinsp;=\u0026thinsp;0.025) demonstrated significant nonlinear dose-response relationships. By contrast, although LDL-C, AC, CRI-I, and CRI-II showed significant overall associations, their tests for nonlinearity were not statistically significant, suggesting that their relationships with the hemorrhagic phenotype were predominantly approximately linear and positive. Overall, the strength of association with the hemorrhagic phenotype increased with increasing levels of cholesterol-related traditional and nontraditional lipid parameters (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eDiscriminative and Incremental Predictive Value of Traditional and Nontraditional Lipid Parameters for the Hemorrhagic Phenotype of MMD\u003c/b\u003e \u003c/p\u003e \u003cp\u003eROC analysis showed that among the traditional lipid parameters, TC and LDL-C had modest discriminative ability for the hemorrhagic phenotype of MMD, with AUCs of 0.637 (95% CI: 0.602\u0026ndash;0.671) and 0.633 (95% CI: 0.598\u0026ndash;0.669), respectively. Among the nontraditional lipid parameters, non-HDL-C performed best, with an AUC of 0.624 (95% CI: 0.589\u0026ndash;0.659), whereas the AUCs of the other markers were generally lower (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Based on these results, TC, LDL-C, non-HDL-C, and CRI-II were further added individually to the basic clinical model to evaluate their incremental predictive value. All four markers improved model discrimination, with the largest increase observed for LDL-C, which improved the C statistic from 0.669 to 0.710. At the same time, TC, LDL-C, and non-HDL-C all showed relatively stable improvements in reclassification, with NRIs of 0.394 (95% CI: 0.264\u0026ndash;0.523), 0.408 (95% CI: 0.274\u0026ndash;0.559), and 0.389 (95% CI: 0.279\u0026ndash;0.524), and corresponding IDIs of 0.0312 (95% CI: 0.0140\u0026ndash;0.0507), 0.0349 (95% CI: 0.0146\u0026ndash;0.0581), and 0.0294 (95% CI: 0.0164\u0026ndash;0.0525), respectively (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). DeLong testing showed that all four lipid-enhanced models were significantly superior to the basic model. Further pairwise comparisons indicated that the LDL-C-enhanced and non-HDL-C-enhanced models both had significantly better discrimination than the CRI-II-enhanced model, whereas the TC-enhanced model did not differ significantly from either the LDL-C-enhanced or the non-HDL-C-enhanced model. Overall, LDL-C and non-HDL-C showed relatively better incremental predictive performance, whereas the overall discriminative performance of TC, LDL-C, and non-HDL-C was similar (Supplementary Table\u0026nbsp;2).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eIncremental predictive value of lipid parameters added to the basic clinical model\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC statistics\u003c/p\u003e \u003cp\u003eEstimate (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNRI\u003c/p\u003e \u003cp\u003eEstimate (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eP value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eIDI\u003c/p\u003e \u003cp\u003eEstimate (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eP value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBasic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.669 (0.635\u0026ndash;0.704)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBasic model\u0026thinsp;+\u0026thinsp;TC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.706 (0.673\u0026ndash;0.739)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.394 (0.264\u0026ndash;0.523)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0312 (0.0140\u0026ndash;0.0507)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBasic model\u0026thinsp;+\u0026thinsp;LDL-C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.710 (0.677\u0026ndash;0.743)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.408 (0.274\u0026ndash;0.559)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0349 (0.0146\u0026ndash;0.0581)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBasic model\u0026thinsp;+\u0026thinsp;Non-HDL-C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.706 (0.674\u0026ndash;0.739)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.389 (0.279\u0026ndash;0.524)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0294 (0.0164\u0026ndash;0.0525)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBasic model\u0026thinsp;+\u0026thinsp;CRI-II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.694 (0.660\u0026ndash;0.727)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.219 (0.085\u0026ndash;0.334)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0185 (0.0073\u0026ndash;0.0337)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.027\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBasic model\u0026thinsp;+\u0026thinsp;AC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.687 (0.654\u0026ndash;0.721)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.180 (0.085\u0026ndash;0.351)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0138 (0.0032\u0026ndash;0.0310)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.052\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBasic model\u0026thinsp;+\u0026thinsp;CRI-I\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.687 (0.654\u0026ndash;0.721)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.180 (0.051\u0026ndash;0.344)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0138 (0.0024\u0026ndash;0.0276)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBasic model\u0026thinsp;+\u0026thinsp;LCI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.677 (0.642\u0026ndash;0.711)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.197 (0.034\u0026ndash;0.321)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.012\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0042 (0.0000-0.0197)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.413\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003eNote: The basic model included sex, age, BMI, hypertension, diabetes, smoking status, drinking status, bilateral involvement, and dichotomized Suzuki stage (\u0026thinsp;\u0026lt;\u0026thinsp;=\u0026thinsp;4 vs\u0026thinsp;\u0026gt;\u0026thinsp;4). Missing BMI values were handled by multiple imputation (5 imputations) before model fitting. C statistics were reported as AUCs with 95% confidence intervals based on the nonparametric DeLong method using pooled predicted probabilities. Continuous NRI and IDI 95% confidence intervals and P values were obtained by bootstrap resampling (100 resamples).\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eSensitivity and Subgroup Analyses\u003c/h2\u003e \u003cp\u003eSensitivity analyses showed that, after repeating the analyses in the complete-case dataset, the main findings remained largely consistent with those of the primary analysis after multiple imputation. TC, LDL-C, non-HDL-C, AC, CRI-I, CRI-II, and LCI remained significantly associated with hemorrhagic MMD, whereas TG, HDL-C, AIP, RC, and RC/HDL-C still showed no stable associations (Supplementary Table\u0026nbsp;3). After further subdividing the non-hemorrhagic group into asymptomatic and ischemic subgroups, the comparison between the hemorrhagic and ischemic groups was highly consistent with the main analysis (Supplementary Table\u0026nbsp;5), whereas the comparison with the asymptomatic group still showed significant associations for TC, LDL-C, and non-HDL-C, although the effects of some markers were attenuated (Supplementary Table\u0026nbsp;4). Subgroup analyses further showed that TC, LDL-C, and non-HDL-C remained positively associated with hemorrhagic MMD across strata of age, sex, hypertension status, and BMI, and none of the interaction tests reached statistical significance, supporting the robustness of these associations (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e; Supplementary Table\u0026nbsp;6).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eMediation Analysis\u003c/h2\u003e \u003cp\u003eTo further investigate the potential mediating role of BMI in the associations between lipid parameters and the hemorrhagic phenotype of MMD, mediation analyses were performed separately for each traditional and nontraditional lipid parameter. The analyses showed that BMI exerted significant mediation effects for some lipid parameters, predominantly in a negative direction. Specifically, significant negative indirect effects were observed for non-HDL-C, AC, CRI-I, and CRI-II. For non-HDL-C, the indirect effect was beta = -0.0370 (P\u0026thinsp;=\u0026thinsp;0.014), with a proportion mediated of -9.40% (95% CI: -17.7% to -1.1%). For AC and CRI-I, the indirect effects were both beta = -0.0644 (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and the proportions mediated were both \u0026minus;\u0026thinsp;30.77% (95% CI: -55.8% to -5.7%). For CRI-II, the indirect effect was beta = -0.0799 (P\u0026thinsp;=\u0026thinsp;0.001), with a proportion mediated of -26.14% (95% CI: -46.8% to -5.5%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e; Supplementary Table\u0026nbsp;7). By contrast, no significant mediation effects were observed for the remaining lipid parameters. These findings suggest that BMI may act more as a suppressive mediator than a conventional concordant mediator in the associations between key nontraditional lipid parameters and the hemorrhagic phenotype of MMD.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis dual-center cross-sectional study of adults with primary MMD systematically evaluated the associations between traditional and nontraditional lipid parameters and the hemorrhagic phenotype, and further examined their dose-response patterns, discriminative and incremental predictive value, and the potential mediating role of BMI. The main findings were as follows. First, after full adjustment for confounding factors, TC, LDL-C, non-HDL-C, AC, CRI-I, CRI-II, and LCI remained significantly associated with the hemorrhagic phenotype, with TC, LDL-C, and non-HDL-C showing the most stable associations and the largest effect sizes. Second, RCS analysis suggested nonlinear dose-response associations of TC, non-HDL-C, and LCI with the hemorrhagic phenotype, and the addition of TC, LDL-C, and non-HDL-C to the basic clinical model improved both model discrimination and reclassification. Third, BMI showed a significant negative mediating effect in the associations between several key lipid parameters and the hemorrhagic phenotype. Overall, these findings suggest that the hemorrhagic phenotype of adult primary MMD may correspond to a lipid metabolic pattern characterized by cholesterol-related lipoprotein abnormalities and may provide supplementary information for risk stratification.\u003c/p\u003e \u003cp\u003eOne of the most important findings of the present study is the relatively consistent and stable association between cholesterol-related lipid markers and the hemorrhagic phenotype of MMD. Compared with TG, RC, RC/HDL-C, and AIP, TC, LDL-C, and non-HDL-C showed stronger and more stable statistical signals across baseline comparisons, multivariable logistic regression, RCS analysis, and incremental predictive analysis. In the fully adjusted model, the ORs for the highest versus the lowest quartiles of TC, LDL-C, and non-HDL-C were all close to threefold, indicating a fairly substantial association with the hemorrhagic phenotype. Although composite indices such as AC, CRI-I, CRI-II, and LCI also remained significant, they did not clearly outperform these three core markers. Taken together, in the special cerebrovascular context of MMD, the lipid abnormalities that are more stably linked to the hemorrhagic phenotype may primarily reflect an increased burden of cholesterol-related lipoproteins and imbalance in lipoprotein composition, rather than isolated hypertriglyceridemia or remnant cholesterol abnormalities.\u003c/p\u003e \u003cp\u003eThis finding should be interpreted in light of the disease-specific pathophysiological background of MMD. In adult MMD, particularly hemorrhagic MMD, the core pathology is not classic large-artery atherosclerosis, but rather progressive stenosis or occlusion of the terminal internal carotid arteries and proximal branches, accompanied by the development of abnormal collateral vascular networks, hemodynamic redistribution, and dysregulated vascular wall remodeling[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Previous AHA/ASA scientific statements and related reviews have pointed out that hemorrhagic MMD in adults is closely associated with fragile collateral circulation and vulnerable periventricular anastomoses, which are regarded as important structural bases for hemorrhage[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. More recent studies have further shown that periventricular anastomoses are not only an important source of hemorrhage but are also closely related to the risk of rebleeding[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Therefore, in this disease setting, metabolic factors that affect vascular wall stability and fragility may have clinical implications distinct from those in general atherosclerotic disease.\u003c/p\u003e \u003cp\u003eAgainst this background, the increased burden of cholesterol-related lipid parameters observed in the present study may not simply represent conventional atherosclerotic risk factors, but may instead indicate a metabolic milieu that promotes vascular wall injury. Previous studies have shown that the apoB-containing atherogenic lipoprotein particles represented by LDL-C and non-HDL-C more directly reflect the burden of injurious lipid exposure to the vascular wall than TG-related markers. Under oxidative and inflammatory conditions, these lipoproteins may reduce nitric oxide bioavailability, increase reactive oxygen species generation, and upregulate inflammatory and adhesion molecules, thereby impairing endothelial homeostasis and weakening vascular wall integrity[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. In patients with MMD, who already have abnormal collateral formation, increased hemodynamic burden, and vascular wall fragility[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e], such a cholesterol-related vascular injury background may further aggravate local vascular instability. In addition, oxidative stress may affect cerebral microvascular endothelial tight junctions and blood-brain barrier stability, thereby increasing the vulnerability of fragile collateral vessels[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Although these mechanisms have not yet been directly confirmed in MMD[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e], general biological mechanisms of cerebrovascular injury suggest that they may represent an important link between cholesterol-related lipid burden and the hemorrhagic phenotype[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. This interpretation is also consistent with the direction of recent studies on lipid metabolism in MMD. One study reported a broader spectrum of lipid abnormalities in MMD, including elevated oxLDL, sdLDL, and Lp(a), with oxLDL showing a relatively strong independent risk signal, suggesting that lipid metabolic abnormalities may participate in the development and progression of MMD[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe present findings also suggest that not all lipid parameters carry equal value for the hemorrhagic phenotype of MMD. TG, RC, RC/HDL-C, and AIP did not show stable associations after full adjustment, whereas TC, LDL-C, and non-HDL-C remained consistently significant. This difference suggests that the metabolic background corresponding to the hemorrhagic phenotype of adult MMD may differ from the classic metabolic syndrome or insulin-resistance phenotype dominated by hypertriglyceridemia[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e], and may instead be more closely related to the vascular injury pattern reflected by cholesterol-related lipoprotein abnormalities. At the same time, the observation that AIP was lower in the hemorrhagic group at baseline is also noteworthy. AIP primarily reflects the relationship between TG and HDL-C[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]; in the present study, the hemorrhagic group did not have substantially higher TG but did have relatively higher HDL-C, making its direction different from those of TC, LDL-C, and non-HDL-C statistically understandable. In addition, higher HDL-C concentrations do not necessarily indicate a parallel enhancement of anti-inflammatory, antioxidant, and endothelial-protective functions, because HDL may become dysfunctional under inflammatory or oxidative stress conditions[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eRCS analysis and incremental predictive analysis further support the notion that TC, LDL-C, and non-HDL-C are more informative than other lipid parameters for characterizing the hemorrhagic phenotype of MMD. RCS analysis showed significant nonlinear associations of TC, non-HDL-C, and LCI with the hemorrhagic phenotype, whereas LDL-C, AC, CRI-I, and CRI-II mainly showed approximately linear positive associations, suggesting that different lipid parameters may have threshold effects or accelerated risk ranges in relation to the hemorrhagic phenotype. At the same time, ROC and incremental analyses showed that although the individual discriminative ability of TC, LDL-C, and non-HDL-C was only moderate, the model AUC increased from 0.669 to 0.706\u0026ndash;0.710 after the inclusion of these markers in the basic clinical model, with statistically significant improvements in both NRI and IDI; among them, LDL-C showed the largest incremental gain. For MMD, a disease that lacks mature biomarkers[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], readily available lipid measures obtained at admission, with low cost and high reproducibility, hold practical translational potential.\u003c/p\u003e \u003cp\u003eSensitivity and subgroup analyses further support the stability of the observed associations. The consistency between complete-case analyses and analyses after multiple imputation suggests that the main conclusions do not depend on the handling of missing BMI values. After the non-hemorrhagic group was further divided into asymptomatic and ischemic subgroups, the results comparing the hemorrhagic and ischemic groups were highly consistent with those of the main analysis, whereas some marker effects were attenuated in comparisons with the asymptomatic group. This pattern suggests that the lipid differences identified in the present study may more likely reflect biological divergence between hemorrhagic and ischemic phenotypes. In addition, TC, LDL-C, and non-HDL-C remained directionally consistent across strata of age, sex, hypertension, and BMI, and no significant interactions were observed, indicating that these associations were robust across clinical subgroups.\u003c/p\u003e \u003cp\u003eThe present study also found that BMI had a negative mediating effect in the associations between several key lipid parameters and the hemorrhagic phenotype. This pattern is inconsistent with the classic pathway in conventional cardiovascular research, in which higher BMI increases event risk by aggravating lipid metabolic disturbances[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. Instead, it is more compatible with a suppression effect: lipid parameters were positively associated with BMI, whereas BMI was negatively associated with the hemorrhagic phenotype of MMD, such that the BMI-related pathway statistically attenuated the overall effects of lipid parameters on the hemorrhagic phenotype[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Several large cohort studies have also reported that lower BMI is associated with an increased risk of hemorrhagic stroke[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. In addition, previous studies have shown that individuals with lower BMI are more prone to cerebral microbleeds[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e], which are regarded as imaging markers of cerebral small-vessel fragility and may, to some extent, resemble the hemorrhagic pattern of newly formed vessels in MMD. In the baseline characteristics of the present study, patients in the hemorrhagic group had lower BMI but higher levels of several cholesterol-related lipid parameters. This finding suggests that the meaning of BMI in MMD may differ from that of obesity exposure in the general population and may, to some extent, reflect nutritional reserve, vascular repair capacity, or vascular fragility.\u003c/p\u003e \u003cp\u003eFrom a clinical translation perspective, the present study has at least three potential implications. First, TC, LDL-C, and non-HDL-C are all derived from routine lipid testing and have the advantages of low cost, wide availability, and good reproducibility[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e], making them potentially useful as supplementary biological information in the admission evaluation of adults with MMD. Second, compared with single lipid parameters, non-HDL-C and several composite markers may better reflect the burden of lipoproteins and imbalance in lipoprotein composition[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Third, given their incremental value beyond the basic clinical model, these lipid parameters could be integrated in the future with imaging risk markers, hemodynamic parameters, and clinical variables to develop a more comprehensive risk assessment model for hemorrhage in adult MMD.\u003c/p\u003e \u003cp\u003eThe present study also has several limitations. First, this was a retrospective dual-center cross-sectional observational study. Although multivariable adjustment, sensitivity analyses, and subgroup analyses were performed, residual confounding cannot be completely excluded, and causal inference cannot be established. Second, the present analysis focused on the hemorrhagic phenotype of adult primary MMD rather than incident hemorrhagic events during prospective follow-up. Therefore, the findings are more applicable to phenotypic differentiation and risk stratification than to prediction of future hemorrhage. Third, all lipid parameters were derived from a single measurement at admission, which cannot reflect long-term lipid exposure or dynamic changes over time and cannot fully rule out the influence of acute events or stress states on metabolic markers. Fourth, although major clinical variables and Suzuki stage were included in the adjustment models, more detailed imaging risk markers and some treatment-related factors, such as prior lipid-lowering therapy, antiplatelet therapy, or more detailed revascularization information, were unavailable. Accordingly, the specific pathways linking lipid parameters with the hemorrhagic phenotype still require further investigation. Future multicenter prospective studies, combined with dynamic lipid monitoring, metabolomics or lipidomics, and vascular wall imaging and cerebral blood flow assessment, are still needed to further clarify the role of cholesterol-related lipid parameters in phenotypic differentiation and hemorrhagic risk assessment in MMD.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, the hemorrhagic phenotype of adult primary MMD was closely associated with elevated cholesterol-related lipid parameters, among which TC, LDL-C, and non-HDL-C showed the most stable associations. The negative mediating effect of BMI further suggests that the relationship between metabolic status and the hemorrhagic phenotype in MMD may differ from that observed in general cardio-cerebrovascular diseases. These findings provide new clues for mechanistic research and risk assessment in the hemorrhagic phenotype of adult primary MMD.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003cstrong\u003eCompeting interests\u003c/strong\u003e \u003cp\u003eThe authors declare no conflicts of interest.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eEthical approval\u003c/strong\u003e \u003cp\u003e This study was approved by the Ethics Committee of Beijing Hospital (2024BJYYEC-KY164-04) and the Ethics Committee of Beijing Tiantan Hospital, Capital Medical University (KY-2022-051-08). The study was conducted in accordance with the Declaration of Helsinki.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent to participate\u003c/strong\u003e \u003cp\u003e Because of the retrospective nature of the study, the requirement for informed consent was waived by the ethics committees.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eZhang Dong, Weihong Huang and Haoyang Xiong conceived and designed the study. Wei Liu, Peicong Ge, Wenjing Chen, Yunhao Liu, Xuefeng Tang, Ziheng Yang, and Zhongji Yan collected the data. Weihong Huang and Wenjing Chen drafted the manuscript. All authors critically revised the manuscript and approved the final version.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe thank the individuals who contributed to the study or manuscript preparation but did not fulfill all the criteria of authorship.\u003c/p\u003e\u003ch2\u003eData availability statement\u003c/h2\u003e\u003cp\u003eThe datasets generated and analyzed during the current study are not publicly available because of patient privacy and institutional restrictions, but are available from the corresponding author on reasonable request.\u003c/p\u003e "},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eDiagnostic. Criteria for Moyamoya Disease \u0026ndash; 2021 Revised Version. Neurol Med Chir (Tokyo). 2022;62(7):307\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2176/jns-nmc.2022-0072\u003c/span\u003e\u003cspan address=\"10.2176/jns-nmc.2022-0072\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoyamoya Disease: Pathophysiology, Diagnosis, and Treatment. Dtsch Arztebl Int. 2025;122(26):722\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3238/arztebl.m2025.0185\u003c/span\u003e\u003cspan address=\"10.3238/arztebl.m2025.0185\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAdult Moyamoya Disease and Syndrome. Current Perspectives and Future Directions: A Scientific Statement From the American Heart Association/American Stroke Association. Stroke. 2023;54(10):e465\u0026ndash;79. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1161/STR.0000000000000443\u003c/span\u003e\u003cspan address=\"10.1161/STR.0000000000000443\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHemorrhagic Moyamoya Disease. A Recent Update. J Korean Neurosurg Soc. 2019;62(2):136\u0026ndash;43. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3340/jkns.2018.0101\u003c/span\u003e\u003cspan address=\"10.3340/jkns.2018.0101\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUnraveling the Dyslipidemic Landscape in Moyamoya Disease: OxLDL as a Key Biomarker. CNS Neurosci Ther. 2025;31(5):e70441. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/cns.70441\u003c/span\u003e\u003cspan address=\"10.1111/cns.70441\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHypo-high density lipoproteinemia is a predictor for recurrent stroke during the long-term follow-up after revascularization in adult moyamoya disease. Front Neurol. 2022;13:891622. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fneur.2022.891622\u003c/span\u003e\u003cspan address=\"10.3389/fneur.2022.891622\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eData-Independent Acquisition-Based Serum Proteomic Profiling of Adult Moyamoya Disease Patients Reveals the Potential Pathogenesis of Vascular Changes. J Mol Neurosci. 2022;72(12):2473\u0026ndash;85. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12031-022-02092-w\u003c/span\u003e\u003cspan address=\"10.1007/s12031-022-02092-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAssociation Between the Onset Pattern of Adult Moyamoya Disease and Risk Factors for Stroke. Stroke. 2020;51(10):3124\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1161/STROKEAHA.120.030653\u003c/span\u003e\u003cspan address=\"10.1161/STROKEAHA.120.030653\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e2021 Japanese Guidelines for the Management of Moyamoya Disease: Guidelines from the Research Committee on Moyamoya Disease and Japan Stroke Society. Neurol Med Chir (Tokyo). 2022;62(4):165\u0026ndash;170. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2176/jns-nmc.2021-0382\u003c/span\u003e\u003cspan address=\"10.2176/jns-nmc.2021-0382\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNonlinear relationship between untraditional lipid parameters and the risk of prediabetes: a large retrospective study based on Chinese adults. Cardiovasc Diabetol. 2024;23(1):12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12933-023-02103-z\u003c/span\u003e\u003cspan address=\"10.1186/s12933-023-02103-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eApoB LDL-C. and non-HDL-C as markers of cardiovascular risk. J Clin Lipidol. 2025;19(4):844\u0026ndash;59. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jacl.2025.05.024\u003c/span\u003e\u003cspan address=\"10.1016/j.jacl.2025.05.024\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePotential Association Between Atherogenic Coefficient. Prognostic Nutritional Index, and Various Obesity Indices in Diabetic Nephropathy. Nutrients. 2025;17(8):1339. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/nu17081339\u003c/span\u003e\u003cspan address=\"10.3390/nu17081339\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDotinurad restores exacerbated kidney dysfunction in hyperuricemic patients with chronic kidney disease. BMC Nephrol. 2024;25(1):97. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12882-024-03535-9\u003c/span\u003e\u003cspan address=\"10.1186/s12882-024-03535-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe lipid ratio castelli's risk index II is a novel biomarker for intraplaque neovascularization in patients with carotid stenosis. Lipids Health Dis. 2025;25(1):11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12944-025-02821-1\u003c/span\u003e\u003cspan address=\"10.1186/s12944-025-02821-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe effects of a novel nutraceutical combination on low-density lipoprotein cholesterol and other markers of cardiometabolic health in adults with hypercholesterolaemia: A randomised double-blind placebo-controlled trial. Atherosclerosis. 2025;403:119177. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.atherosclerosis.2025.119177\u003c/span\u003e\u003cspan address=\"10.1016/j.atherosclerosis.2025.119177\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang X, Li C, Zeng P, Ling Y, Tan S, Bai Z, Shen S, Chen S, Nie B, Wang H, Lyu J. Non-traditional lipid-inflammatory parameters estimate the risk of stroke in middle-aged and older Chinese adults: a nationwide prospective cohort study. J Adv Res 2025 Nov 17:S2090-1232(25)00924\u0026ndash;5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jare.2025.11.029\u003c/span\u003e\u003cspan address=\"10.1016/j.jare.2025.11.029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLipid metabolism. BMI and the risk of nonalcoholic fatty liver disease in the general population: evidence from a mediation analysis. J Transl Med. 2023;21(1):192. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12967-023-04047-0\u003c/span\u003e\u003cspan address=\"10.1186/s12967-023-04047-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e[Adult Hemorrhagic Moyamoya Disease. Physiopathology and Treatment]. No Shinkei Geka. 2025;53(3):576\u0026ndash;82. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.11477/mf.030126030530030576\u003c/span\u003e\u003cspan address=\"10.11477/mf.030126030530030576\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHigh rebleeding risk associated with choroidal collateral vessels in hemorrhagic moyamoya disease: analysis of a nonsurgical cohort in the Japan Adult Moyamoya Trial. J Neurosurg. 2019;130(2):525\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3171/2017.9.JNS17576\u003c/span\u003e\u003cspan address=\"10.3171/2017.9.JNS17576\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe Role of Non-HDL Cholesterol and Apolipoprotein B in Cardiovascular Disease: A Comprehensive Review. J Cardiovasc Dev Dis. 2025;12(7):256. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/jcdd12070256\u003c/span\u003e\u003cspan address=\"10.3390/jcdd12070256\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUnravelling the Mechanisms of Oxidised Low-Density Lipoprotein in Cardiovascular Health: Current Evidence from In Vitro and In Vivo Studies. Int J Mol Sci. 2024;25(24):13292. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijms252413292\u003c/span\u003e\u003cspan address=\"10.3390/ijms252413292\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMedComm. (2020). 2024;5(8):e651. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/mco2.651\u003c/span\u003e\u003cspan address=\"10.1002/mco2.651\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTemporal Base Transdural Anastomosis in Moyamoya Disease. A Potential Association with Posterior Cerebral Artery Involvement and Clinical Importance. Neurol Med Chir (Tokyo). 2025;65(8):366\u0026ndash;72. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2176/jns-nmc.2025-0113\u003c/span\u003e\u003cspan address=\"10.2176/jns-nmc.2025-0113\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe role of oxidative stress in blood-brain barrier disruption during ischemic stroke: Antioxidants in clinical trials. Biochem Pharmacol. 2024;228:116186. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.bcp.2024.116186\u003c/span\u003e\u003cspan address=\"10.1016/j.bcp.2024.116186\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAging. vascular dysfunction, and the blood-brain barrier: unveiling the pathophysiology of stroke in older adults. Biogerontology. 2025;26(2):67. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10522-025-10209-y\u003c/span\u003e\u003cspan address=\"10.1007/s10522-025-10209-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. 2009;120(16):1640\u0026ndash;1645. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1161/CIRCULATIONAHA.109.192644\u003c/span\u003e\u003cspan address=\"10.1161/CIRCULATIONAHA.109.192644\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAdipose Tissue Insulin Resistance. A Key Driver of Metabolic Syndrome Pathogenesis. Biomedicines. 2025;13(10):2376. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/biomedicines13102376\u003c/span\u003e\u003cspan address=\"10.3390/biomedicines13102376\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAtherogenic. index of plasma [log(triglycerides/HDL-cholesterol)]: theoretical and practical implications. Clin Chem. 2004;50(7):1113\u0026ndash;1115. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1373/clinchem.2004.033175\u003c/span\u003e\u003cspan address=\"10.1373/clinchem.2004.033175\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHDL in the 21st Century: A Multifunctional Roadmap for Future HDL Research. Circulation. 2021;143(23):2293\u0026ndash;2309. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1161/CIRCULATIONAHA.120.044221\u003c/span\u003e\u003cspan address=\"10.1161/CIRCULATIONAHA.120.044221\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDysfunctional high-density lipoprotein: an updated review. Front Cardiovasc Med. 2025;12:1713387. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fcvm.2025.1713387\u003c/span\u003e\u003cspan address=\"10.3389/fcvm.2025.1713387\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMolecular. and multimodal biomarkers in Moyamoya disease: from pathogenic mechanisms to clinical translation. Eur J Med Res. 2026;31(1):187. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s40001-025-03769-9\u003c/span\u003e\u003cspan address=\"10.1186/s40001-025-03769-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eObesity and Cardiovascular Disease. A Scientific Statement From the American Heart Association. Circulation. 2021;143(21):e984\u0026ndash;1010. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1161/CIR.0000000000000973\u003c/span\u003e\u003cspan address=\"10.1161/CIR.0000000000000973\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEquivalence of the mediation, confounding and suppression effect. Prev Sci. 2000;1(4):173\u0026ndash;81. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1023/a:1026595011371\u003c/span\u003e\u003cspan address=\"10.1023/a:1026595011371\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReasons for Testing Mediation in the Absence of an Intervention Effect. A Research Imperative in Prevention and Intervention Research. J Stud Alcohol Drugs. 2018;79(2):171\u0026ndash;81. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.15288/jsad.2018.79.171\u003c/span\u003e\u003cspan address=\"10.15288/jsad.2018.79.171\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAdiposity. risk of ischaemic and haemorrhagic stroke in 0.5 million Chinese men and women: a prospective cohort study. Lancet Glob Health. 2018;6(6):e630\u0026ndash;40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S2214-109X(18)30216-X\u003c/span\u003e\u003cspan address=\"10.1016/S2214-109X(18)30216-X\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAdiposity. ischemic and hemorrhagic stroke: Prospective study in women and meta-analysis. Neurology. 2016;87(14):1473\u0026ndash;81. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1212/WNL.0000000000003171\u003c/span\u003e\u003cspan address=\"10.1212/WNL.0000000000003171\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClinical impact of body mass index on outcomes of ischemic and hemorrhagic strokes. Int J Stroke. 2024;19(8):907\u0026ndash;15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/17474930241249370\u003c/span\u003e\u003cspan address=\"10.1177/17474930241249370\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSevere underweight and cerebral microbleeds. J Neurol. 2012;259(12):2707\u0026ndash;13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00415-012-6574-7\u003c/span\u003e\u003cspan address=\"10.1007/s00415-012-6574-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e2026 ACC/AHA/AACVPR/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Dyslipidemia: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2026 Mar 13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1161/CIR.0000000000001423\u003c/span\u003e\u003cspan address=\"10.1161/CIR.0000000000001423\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e[Chinese guidelines for lipid management. (2023)]. Zhonghua Xin Xue Guan Bing Za Zhi. 2023;51(3):221\u0026ndash;255. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3760/cma.j.cn112148-20230119-00038\u003c/span\u003e\u003cspan address=\"10.3760/cma.j.cn112148-20230119-00038\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eResidual Atherosclerotic Cardiovascular Disease Risk. Focus on Non-High-Density Lipoprotein Cholesterol. J Cardiovasc Pharmacol Ther. 2023;28:10742484231189597. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/10742484231189597\u003c/span\u003e\u003cspan address=\"10.1177/10742484231189597\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"translational-stroke-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"trsr","sideBox":"Learn more about [Translational Stroke Research](http://jcmr-online.biomedcentral.com)","snPcode":"12975","submissionUrl":"https://submission.nature.com/new-submission/12975/3","title":"Translational Stroke Research","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"moyamoya disease, hemorrhagic phenotype, lipid parameters, non-HDL cholesterol, risk stratification","lastPublishedDoi":"10.21203/rs.3.rs-9393575/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9393575/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eObjective\u003c/h2\u003e \u003cp\u003eTo investigate the associations of traditional and novel lipid parameters with the hemorrhagic phenotype in adults with primary moyamoya disease (MMD) and to evaluate their potential value for phenotype discrimination and risk stratification.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eIn this retrospective dual-center cross-sectional study, 1,176 adults with primary MMD treated at Beijing Hospital and Beijing Tiantan Hospital between January 2022 and January 2026 were included, comprising 857 patients with a non-hemorrhagic phenotype and 319 with a hemorrhagic phenotype. Traditional and novel lipid parameters were derived from routine lipid measurements. Missing body mass index (BMI) values were handled using multiple imputation. Multivariable logistic regression, restricted cubic spline analysis, receiver operating characteristic analysis, incremental predictive analysis, sensitivity analyses, subgroup analyses, and mediation analysis were performed.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eCompared with patients with the non-hemorrhagic phenotype, those with the hemorrhagic phenotype had lower BMI and higher levels of multiple cholesterol-related lipid parameters. In multivariable analyses, total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and non-high-density lipoprotein cholesterol (non-HDL-C) were significantly associated with the hemorrhagic phenotype, with odds ratios for the highest versus lowest quartile of 3.435 (95% CI, 2.254\u0026ndash;5.237), 3.197 (95% CI, 2.134\u0026ndash;4.791), and 3.150 (95% CI, 2.087\u0026ndash;4.754), respectively. TC, LDL-C, and non-HDL-C showed modest discriminative ability and improved the performance of the basic clinical model. Sensitivity and subgroup analyses were generally consistent. BMI showed a significant negative mediating effect in the associations between several key novel lipid parameters and the hemorrhagic phenotype.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eCholesterol-related traditional and novel lipid parameters, particularly TC, LDL-C, and non-HDL-C, were independently associated with the hemorrhagic phenotype in adult primary MMD and may provide clinically accessible markers for phenotype stratification.\u003c/p\u003e","manuscriptTitle":"Association of Traditional and Novel Lipid Indicators With the Hemorrhagic Phenotype in Adult Moyamoya Disease: Implications for Lipid Risk Stratification","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-06 17:05:06","doi":"10.21203/rs.3.rs-9393575/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-12T06:24:32+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-10T16:26:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"252771275523975046735564933014394856420","date":"2026-04-29T11:39:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"199308680315756738777463452854389344209","date":"2026-04-29T00:07:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"278092963288683020602372985915484826336","date":"2026-04-28T06:14:05+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-28T00:57:11+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-18T18:37:12+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-13T06:58:28+00:00","index":"","fulltext":""},{"type":"submitted","content":"Translational Stroke Research","date":"2026-04-12T10:40:42+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"translational-stroke-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"trsr","sideBox":"Learn more about [Translational Stroke Research](http://jcmr-online.biomedcentral.com)","snPcode":"12975","submissionUrl":"https://submission.nature.com/new-submission/12975/3","title":"Translational Stroke Research","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d6b68c31-468a-4faa-98b8-e3eceab34946","owner":[],"postedDate":"May 6th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-12T06:24:32+00:00","index":18,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-10T16:26:28+00:00","index":17,"fulltext":""},{"type":"reviewerAgreed","content":"252771275523975046735564933014394856420","date":"2026-04-29T11:39:27+00:00","index":16,"fulltext":""},{"type":"reviewerAgreed","content":"199308680315756738777463452854389344209","date":"2026-04-29T00:07:11+00:00","index":15,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-06T17:05:06+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-06 17:05:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9393575","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9393575","identity":"rs-9393575","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}