文献速递神经MR

多模态MRI揭示脑小血管病结构-功能耦合分期重组

Multimodal MRI Reveals Stage-Specific Reorganization of Structure-Function Coupling from Compensatory to Decoupled States in Cerebral Small Vessel Disease.
PubMed 文献 · Acad Radiol · 2026-07-14

该横断面研究利用ADNI数据库中同时具备DTI和静息态fMRI的280名参与者,按CSVD负荷评分分为无、轻、重三组。研究发现轻度组出现岛叶及眶额区结构-功能耦合增强,重度组则表现为广泛结构网络效率下降、海马及罗兰多岛盖耦合减弱,呈分期样改变。

要点

  • 轻度CSVD以岛叶/眶额耦合增强为特征,提示代偿期改变。
  • 重度CSVD出现海马和罗兰多岛盖耦合减低及广泛结构网络损害。
  • 左侧岛叶耦合与MoCA评分负相关,提示功能重组与认知相关。

核心结论

  • 轻度CSVD组表现为岛叶和眶额皮质等区域结构-功能耦合增强,重度组则出现广泛结构网络差异、海马及罗兰多岛盖耦合下降及功能节点改变。
  • CSVD负荷组间的横断面比较揭示与分期样进展一致的重组模式,但横断面设计无法推断因果。

研究设计与样本

  • 横断面研究,数据来源于ADNI数据库,纳入同时具备DTI和静息态fMRI数据的参与者。
  • 共280名参与者,平均年龄75.0±7.9岁,中位76岁,49.6%为女性。
  • 按CSVD负荷评分0-4分分为无CSVD组118例、轻度CSVD组80例、重度CSVD组82例。
  • 结构网络由DTI数据经确定性纤维追踪重建,功能网络由静息态fMRI区域时间序列Pearson相关获得。
  • 采用GRETNA进行图论分析,组间比较使用单因素方差分析,以年龄、性别、教育为协变量,FDR校正。

关键影像表现/征象

  • 重度CSVD组结构网络全局效率降低(7.42±2.63对8.42±2.75,p=0.03),局部效率降低(16.00±4.77对17.83±5.16,p=0.02)。
  • 重度CSVD组结构网络路径长度增加(0.15±0.04对0.13±0.04,p=0.003)。
  • 轻度CSVD组可见左侧海马节点损害(p=0.0008),重度组节点损害广泛(均p<0.001)。
  • 功能网络无全局性差异,重度CSVD组在扣带回/Heschl回出现节点改变(p<0.001)。
  • 轻度组岛叶/眶额区耦合增强(p=0.002-0.041)、苍白球耦合降低(p=0.026);重度组嗅/枕/颞区耦合增强(p=0.001-0.023)、罗兰多岛盖耦合降低(p=0.019)。

诊断标准与关键数值

  • CSVD负荷评分范围0-4分,用于将参与者分为无、轻、重三组。
  • 左侧岛叶耦合与MoCA评分呈负相关(r=-0.245,p=0.015)。
  • 统计显著性设定为双尾p<0.05,节点和网络水平经FDR校正。
  • 重度组海马耦合低于轻度组(p=0.009)。

临床应用与阅片要点

  • 轻度CSVD的岛叶/眶额耦合增强可能代表早期代偿性功能重组,阅片时需关注这些区域的网络变化。
  • 重度CSVD结构网络广泛效率下降和路径增长提示白质完整性损害已影响网络拓扑,不宜只看局部病灶。
  • 海马与罗兰多岛盖耦合下降可作为提示疾病进展的功能网络标志,结合认知评分更有利于综合评估。

局限性

  • 横断面设计无法推断CSVD进展与网络重组之间的因果关系。
  • 研究数据来源于ADNI数据库,可能存在选择偏倚。
  • 图论指标受纤维追踪方法和网络构建阈值影响,结果可重复性需进一步验证。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

RATIONALE AND OBJECTIVES: Cerebral small vessel disease (CSVD) affects white matter integrity and can alter the brain's network architecture. However, the relationship between structural and functional connectivity in CSVD remains underexplored. This study aims to investigate the differences in topological properties of structural and functional brain networks across CSVD burden groups.

MATERIALS AND METHODS: This cross-sectional study utilized data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu), concentrating on participants who had both DTI and resting-state fMRI data. Participants were classified into non-CSVD (CSVD-n), mild CSVD (CSVD-m), or severe CSVD (CSVD-s) groups based on a CSVD burden score ranging from zero to four. Structural networks were reconstructed from DTI data using deterministic tractography, while functional networks were derived from resting-state fMRI data through Pearson's correlation of regional time series. Graph theoretical analyses were conducted using GRETNA in MATLAB to calculate global and nodal metrics, as well as structure-function coupling indices. Group comparisons were performed using one-way ANOVA (one-way analysis of variance), with age, sex, and education as covariates. Statistical significance was assessed at a two-tailed p < 0.05, adjusted for multiple comparisons across nodes and networks using false discovery rate (FDR) correction.

RESULTS: The study included 280 participants (mean 75.0±7.9y, median 76y; 49.6% female). The participants were divided into groups: CSVD‑n (118, 42.1%), CSVD‑m (80, 28.6%), and CSVD‑s (82, 29.3%). Analysis of structural networks revealed that CSVD‑s exhibited reduced global/local efficiency (7.42±2.63 vs. 8.42±2.75, p = 0.03; 16.00±4.77 vs. 17.83±5.16, p = 0.02) and increased path length (0.15±0.04 vs. 0.13±0.04, p = 0.003). Nodal damage was observed in the left hippocampus in CSVD‑m (p = 0.0008) and was widespread in CSVD‑s (all p < 0.001). In terms of functional networks, no global differences were found (all p > 0.05); however, CSVD‑s exhibited nodal changes in the cingulate/Heschl gyrus (p < 0.001). Regarding coupling, CSVD‑m showed increased coupling in the insula/orbitofrontal regions (p = 0.002-0.041) and decreased coupling in the pallidum (p = 0.026). Conversely, CSVD‑s demonstrated increased coupling in the olfactory/occipital/temporal regions (p = 0.001-0.023) and decreased coupling in the rolandic operculum (p = 0.019) and hippocampus (compared to CSVD‑m, p = 0.009). Furthermore, left insular coupling was correlated with MoCA (r=-0.245, p = 0.015).

CONCLUSION: In the mild CSVD group, researchers observed increased coupling in regions such as the insula and orbitofrontal cortex. In contrast, the severe CSVD group exhibited widespread structural network differences, decreased coupling in the hippocampus and rolandic operculum, and functional nodal alterations. These cross-sectional comparisons among CSVD burden groups reveal a pattern of differences consistent with stage-like progression. However, causality cannot be inferred from this cross-sectional design.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Wang, Lin, Xu, et al. Multimodal MRI Reveals Stage-Specific Reorganization of Structure-Function Coupling from Compensatory to Decoupled States in Cerebral Small Vessel Disease. Acad Radiol. 2026. PMID: 42448482. doi:10.1016/j.acra.2026.06.044

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。