分享:
分享到微信朋友圈
X
临床研究
急性轻度创伤性脑损伤患者的个体形态学脑网络研究
严家豪 黄文静 王俊 李岩 熊钓寒 张静

Cite this article as YAN J H, HUANG W J, WANG J, et al. Individual‐based morphological brain network and its association with acute mild traumatic brain injury[J]. Chin J Magn Reson Imaging, 2024, 15(5): 55-60.本文引用格式严家豪, 黄文静, 王俊, 等. 急性轻度创伤性脑损伤患者的个体形态学脑网络研究[J]. 磁共振成像, 2024, 15(5): 55-60. DOI:10.12015/issn.1674-8034.2024.05.010.


[摘要] 目的 探究急性轻度创伤性脑损伤(mild traumatic brain injury, mTBI)患者个体形态学脑网络的拓扑属性改变。材料与方法 共纳入43例mTBI患者和37例健康对照(health control, HC),采集所有受试者高分辨率T1WI的图像,使用Freesurfer软件对数据进行预处理得到5个形态学指标(皮层厚度、灰质体积、脑表面积、脑沟深度、平均曲率)脑图并构建个体形态学脑网络,使用图论分析的方法计算网络拓扑属性,通过双样本t检验比较组间差异,并采用错误发生率(false discovery rate, FDR)对结果进行多重比较校正。结果 与HC组相比,mTBI组左侧额极横回(t=-2.186,P=0.032)、中央沟(t=-2.617,P=0.011)、外侧沟水平支(t=-2.456,P=0.016)和右侧枕极(t=-2.013,P=0.048)的节点度中心性(degree centrality, DC)增高,左侧额极横回(t=-2.182,P=0.032)、中央沟(t=-2.226,P=0.029)、外侧沟水平支(t=-2.440,P=0.017)和右侧楔前叶(t=-2.207,P=0.030)的节点效率(nodal efficiency, Ne)增高,与认知和执行功能相关;左侧楔叶(t=2.173,P=0.033)、角回(t=2.498,P=0.015)、海马旁回(t=4.009,P<0.001)的节点DC下降,左侧颞上回极平面(t=2.394,P=0.019)、角回(t=2.668,P=0.009)、海马旁回(t=4.671,P<0.001)、胼胝体沟(t=2.189,P=0.032)的Ne下降,与记忆和情绪调节相关。对于全局拓扑属性,二者差异无统计学意义(P>0.05)。结论 急性期mTBI的个体形态学脑网络仍保持了小世界属性。mTBI异常增高的节点DC和Ne,主要集中在与认知和执行功能相关的脑区,反映了大脑对认知功能的应激性代偿;异常减低的节点DC和Ne,主要集中在与记忆、情绪调节相关的脑区,揭示了急性期的认知和情绪变化。这为急性期mTBI的研究提供了新的角度,对其大脑网络改变机制的探索提供了进一步线索。
[Abstract] Objective To explore the changes of individual-based morphological brain network topological properties in patients with mild traumatic brain injury (mTBI).Materials and Methods A sample of 43 mTBI patients and 37 healthy controls (HC) were included. After T1WI data of all subjects were collected, using the Freesurfer to do data preprocessing and geting severals morphological indices (cortical thickness, gray matter volume, brain surface area, sulcus depth, mean curvature), which construct the individual-based morphological brain network, and then Graph Theoretical Network Analysis (GRETNA) was utilized to calculate the network topological properties. The last, differences between the two groups were compared based on two-samples t-test, with false discovery rate (FDR) corrections for multiple comparisons.Results Compared with the HC group, the mTBI group showed approximately differences in nodal degree centrality (DC), the brain regions of left frontal pole transverse gyrus (t=-2.186, P=0.032), central sulcus (t=-2.617, P=0.011), horizontal anterior segment of lateral fissure (t=-2.456, P=0.016) and right pole occipital (t=-2.013, P=0.048) were increased, and nodal efficiency (Ne) of left frontal pole transverse gyrus (t=-2.182, P=0.032;), central sulcus (t=-2.226, P=0.029), horizontal anterior segment of lateral fissure (t=-2.440, P=0.017) and right precuneus (t=-2.207, P=0.030) were increased, which were related to cognitive and executive functions; in addition, the memory and emotional regulation cortexs suffered biggest drop, the DC of the nodes in the left cuneus (t=2.173, P=0.033), angular return (t=2.498, P=0.015) and parahippocampal gyrus (t=4.009, P<0.001) were dropped, and Ne in the left planum polare of the superior temporal gyrus (t=2.394, P=0.019), angular return (t=2.668, P=0.009), parahippocampal gyrus (t=4.671, P<0.001), callosal sulcus (t=2.189, P=0.032) were dropped. The global topological properties differences were not statistically significant (P>0.05).Conclusions The individual morphological brain network of mTBI in the acute phase still retains the small-world attribute. The abnormally elevated nodes DC and Ne of mTBI are mainly concentrated in brain regions related to cognitive and executive functions, reflecting the brain's stress compensation for cognitive functions. The abnormally reduced nodes DC and Ne, mainly concentrated in brain regions related to memory and emotion regulation, revealed cognitive and emotional changes in the acute phase. This provides a new perspective for the study of mTBI in the acute phase and provides further clues for the exploration of the mechanism of brain network alteration.
[关键词] 轻度创伤性脑损伤;结构磁共振成像;磁共振成像;形态学脑网络;图论分析
[Keywords] mild traumatic brain injury;structural magnetic resonance imaging;magnetic resonance imaging;morphological brain network;graph theory analysis

严家豪 1, 2, 3   黄文静 1, 2, 3   王俊 1, 2, 3   李岩 4   熊钓寒 1, 2, 3   张静 1, 2, 3*  

1 兰州大学第二医院核磁共振科,兰州 730030

2 兰州大学第二临床学院,兰州 730000

3 甘肃省功能及分子影像临床医学研究中心,兰州 730030

4 兰州大学数学与统计学院,兰州 730000

通信作者:张静,E-mail:lztong2001@163.com

作者贡献声明::张静设计本研究的方案,对稿件重要的智力内容进行了修改,获得了甘肃省科技计划项目的资助;严家豪起草和撰写稿件,获取、分析和解释本研究的数据;黄文静、王俊、李岩、熊钓寒获取、分析或解释本研究的数据,对稿件重要的智力内容进行了修改;全体作者都同意发表最后的修改稿,同意对本研究的所有方面负责,确保本研究的准确性和诚信。


基金项目: 甘肃省科技计划项目 21JR7RA438,21JR7RA403
收稿日期:2023-08-23
接受日期:2024-03-04
中图分类号:R445.2  R651.15  R395.2 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2024.05.010
本文引用格式严家豪, 黄文静, 王俊, 等. 急性轻度创伤性脑损伤患者的个体形态学脑网络研究[J]. 磁共振成像, 2024, 15(5): 55-60. DOI:10.12015/issn.1674-8034.2024.05.010.

0 引言

       随着交通工具使用率的增加和人口老龄化程度加深,因车祸和跌倒发生轻度创伤性脑损伤(mild traumatic brain injury, mTBI)的事件逐年增加,据估计全球每年有超过八百万人因mTBI住院[1]。超过15%的mTBI患者在受伤后会出现认知和情绪障碍,给患者家庭和社会带来极大的负担[2]。但急性期mTBI患者的常规影像学检查多表现为正常,并且创伤后临床症状多无特异性,导致临床诊断困难。mTBI是导致轻度认知障碍、痴呆和阿尔茨海默病的重要危险因素[3]。因此,对mTBI患者的认知和情绪障碍的早期诊断和及时干预治疗具有重要的临床意义。

       T1WI作为临床常规序列,拥有高分辨率、预处理简单、噪声少等优势。近年来,利用T1WI构建组水平的形态学脑网络已被广泛用于研究神经退行性疾病的脑结构变化,它表征了大脑皮层厚度、体积、表面积等形态学指标的脑区间相似性,反映了解剖区域之间的发育协调性[4]。组水平的形态学脑网络通过计算所有受试者脑区之间平均皮尔逊相关系数获得大脑皮层的形态学关系,只能为一组特定人群构建一个网络。该方法适用于受试者较多的研究,但忽略了个体间异质性,不足以充分揭示可能存在的神经生物学机制[5]。而个体形态学脑网络为每个受试者构建一个网络,保留了个体间的异质性,已成功用于研究多种神经系统疾病,如癫痫、精神病、孤独症谱系障碍等[6, 7, 8]。该方法在帮助临床诊断和预后方面表现出很好的潜在应用前景[9]。目前,尚不清楚急性期mTBI患者个体形态学脑网络将如何变化,本研究拟采用KONG等[10]提出的利用KL散度(Kullback-Leibler divergence, KLD)估计脑区间形态相似性的方法对急性期mTBI患者进行研究,并利用图论分析的方法探究其个体形态学脑网络拓扑属性的变化,以期为急性期mTBI患者的神经病理机制提供更全面的发现,为其早期诊断提供客观神经影像标记。

1 材料与方法

1.1 一般资料

       前瞻性纳入2021年9月至2023年2月就诊于兰州大学第二医院、符合纳入标准的mTBI患者为病例组。纳入标准:(1)患者于就诊前7天内经历过脑部外伤。(2)符合以下2002年美国康复医学会对mTBI的定义至少一条。①意识丧失<30分钟;②创伤后失忆<24小时;③24小时后对患者进行格拉斯昏迷评分量表(Glasgow Coma Scale, GCS)评分,评分≥13分;④创伤后患者的精神状态差,出现眩晕,失去辨别方向的能力;⑤患者可能伴随神经功能缺损。(3)在常规MRI检查时未发现明显的创伤所致的脑损伤。(4)年龄为18~60岁。(5)右利手。排除标准:(1)合并颅骨骨折、脑挫裂伤、胸腔或腹腔损伤、脊髓损伤等;(2)既往有严重的全身疾病病史、精神神经病史、开颅手术史等;(3)长期服用抗精神类药、止痛药、酗酒等。

       同期纳入健康、自愿且符合纳排标准的人群作为健康对照(healthy controls, HC)组,保证两组的年龄、性别、受教育程度相仿。入选标准:(1)既往身体健康,无脑外伤史、手术史,无长期服用药物史;(2)GCS评分15分;(3)右利手。排除标准:MRI检查发现假牙伪影过重及其他局灶性信号异常。

       本研究遵守《赫尔辛基宣言》,经兰州大学第二医院伦理委员会批准,批准文号:2021A-219,全体受试者均签署了知情同意书。

1.2 图像采集和预处理

       所有受试者均在3.0 T MR扫描仪(Siemens Verio,Erlangen,Germany)下完成检查,8通道相位控阵线圈扫描头颅。检查时受试者采用仰卧位,双侧耳旁放置海绵垫减少噪音并固定头部。T1WI扫描采用磁化准备快速采集梯度回波(magnetization-prepared rapid acquisition gradient-echo, MPRAGE)序列:层厚1 mm,层数 192,TR 1900 ms,TE 2.93 ms,TI 900 ms,FA 9°,FOV 256 mm×256 mm,矩阵256×256。并通过T2WI、T2液体衰减反转恢复(T2-fluid attenuated inversion recovery, T2-FLAIR)序列、扩散加权成像(diffusion weighted imaging, DWI)、磁敏感加权成像(susceptibility weighted imaging, SWI)序列排除脑挫裂伤、微出血灶、脑软化灶等。

       使用Freesurfer 6.0(Massachusetts General Hospital,Harvard Medical School,http://surfer.nmr.mgh.harvard.edu)对所有结构图像进行预处理,获得5个基于表面的顶点形态学指标(皮层厚度、灰质体积、脑表面积、脑沟深度、平均曲率)脑图。由两名具有20年和15年临床经验的核磁共振科主任医师和副主任医师通过双盲评估影像结果,对于存在问题的图像(如伪影和不准确的分割)进行Kappa一致性检验,确保诊断的客观性和重复性。

1.3 个体形态学脑网络的构建

       使用基于KLD相似性(KLD-based similarity, KLDs)及其变体JS散度相似性(Jensen-Shannon divergence-based similarity, JSDs)的方法,对每个受试者生成的形态学指标脑图构建形态学脑网络,见图1。网络节点使用Destrieux图谱(a2009s图谱),将双侧大脑半球分为148个皮质脑区,脑区间的相似性表示网络的边。利用KLDs和JSDs量化两个脑区间的相似性。基于核密度估计(kernel density estimation, KDE)计算每个脑区的各形态学指标的概率密度估计值。所得的概率密度函数(每个脑区随机选100个)进一步转换为概率分布函数(probability density function, PDF)。KLD是非对称性的,即KLD(P||Q)和KLD(Q||P)不相等,而我们需要一个对称性的度量指标,因此,基于各形态学指标的PDF脑区P和脑区Q之间的KLDs和JSDs计算公式为:

       其中e为自然指数,n为采样点总数(n=27),

       KLDs和JSDs的取值范围为[0,1],0表示两概率分布的最大可分离性,1表示两概率分布完全相同[11]

图1  本研究的成像处理、网络构建、拓扑表征和统计分析的流程图。①对于mTBI组和HC组中的结构图像,提取基于每个顶点的皮层厚度图。②根据a2009s图谱将皮层厚度图划分为148个脑区。③根据大脑皮层厚度的信号分布估计每个脑区的概率分布函数,并分别用于估计基于KLDs和JSDs的脑区间相似性。④每个受试者共生成10个相似矩阵(5个形态学指标×2个相似性度量指标)。⑤使用基于稀疏性的程序将每个矩阵阈值化为二值网络。最后,计算基于图论分析的7个全局属性和2个节点属性。mTBI:轻度创伤性脑损伤;HC:健康对照;KLDs:Kullback-Leibler散度相似性;JSDs:Jensen-Shannon散度相似性。
Fig. 1  Flow chart of imaging processing, network construction, topological characterization and statistical analysis in this study. ① Extract cortical thickness maps based on each vertex for the structural images in the mTBI and HC groups. ② According to the a2009s atlas, the cortical thickness map is divided into 148 brain regions. ③ The probability distribution function is estimated for each region in terms of signal distribution of cortical thickness maps and was used to estimate interregional similarity with JSDs and KLDs, respectively. ④ This forms a total of 10 similarity matrices for each image (5 morphological indices × 2 similarity smeasures). ⑤ A sparsity-based procedure is further used to threshold each of them into a series of binary networks. Finally, 7 global and 2 nodal graph theory-based network metrics were calculated to characterize topological organization of each binary network. mTBI: mild traumatic brain injury; HC: healthy control; KLDs: Kullback-Leibler divergence-based similarity; JSDs: Jensen-Shannon divergence-based similarity.

1.4 网络分析

       使用GRETNA 2.0(Graph Theoretical Network Analysis,https://www.nitrc.org/projects/gretna)软件进行网络拓扑属性分析,首先对个体形态学脑网络进行二值化处理,网络稀疏度阈值的取值范围为0.034~0.400,步长为0.02,保证网络同时具有小世界属性和稀疏特性,分别计算mTBI组与HC组在该范围内所有19个阈值点的大脑网络拓扑属性:全局指标包括聚类系数(clustering coefficient, Cp)、标准化Cp(gamma, γ)、特征路径长度(the shortest path length, Lp)、标准化Lp(lambda, λ)、小世界属性(sigma, σ)、全局效率(global efficiency, Eglob)、局部效率(local efficiency, Eloc);节点指标包括节点度中心性(degree centrality, DC)、节点效率(nodal efficiency, Ne)。

1.5 统计学分析

       采用SPSS 27.0软件进行统计分析。双样本t检验比较mTBI组与HC组年龄、受教育年限的组间差异,卡方检验比较两组间性别构成的差异,P<0.05为差异具有统计学意义。全局指标和节点指标的组间比较使用双样本t检验。使用曲线下面积(area under the curve, AUC)值分析每个拓扑属性的组间差异时,以性别、年龄、受教育年限为协变量,并采用错误发生率(false discovery rate, FDR)对结果进行多重比较校正,校正后P<0.05具有统计学意义。最后,将计算得到的差异有统计学意义的脑区用BrainNet Viewer(https://www.nitrc.org/projects/bnv)软件进行可视化。

2 结果

2.1 人口统计学分析

       mTBI组与HC组在性别、年龄及受教育年限方面差异均无统计学意义(P>0.05),见表1。mTBI组GCS评分为(14.44±0.77)分,>13分。

表1  两组的人口学资料
Tab. 1  Demographic data of two groups

2.2 全局拓扑属性分析

       在各稀疏度阈值下的AUC分析证明,mTBI组和HC组各形态学指标的全局属性组间差异均无统计学意义(P>0.05,FDR校正)。两组的形态学脑网络均具有小世界属性,即Sigma>1、Lambda≈1和Gamma>1,见图2

图2  形态学脑网络的全局属性指标随稀疏度变化曲线图。2A:基于KLDs的mTBI组;2B:基于KLDs的HC组;2C:基于JSDs的mTBI组;2D:基于JSDs的HC组。mTBI:轻度创伤性脑损伤;HC:健康对照;KLDs:Kullback-Leibler散度相似性;JSDs:Jensen-Shannon散度相似性;Eglob:全局效率;Eloc:局部效率;Cp:聚类系数;Lp:特征路径长度;γ:标准化Cp;λ:标准化Lp;σ:小世界属性。
Fig. 2  The curve of the global attribute index of morphological brain network with sparsity. 2A: mTBI group based on KLDs; 2B: HC group based on KLDs; 2C: mTBI group based on JSDs; 2D: HC group based on JSDs. mTBI: mild traumatic brain injury; HC: healthy control; KLDs: Kullback-Leibler divergence-based similarity; JSDs: Jensen-Shannon divergence-based similarity; Eglob: global efficiency; Eloc: local efficiency; Cp: clustering coefficient; Lp: the shortest path length; γ: standardized Cp; λ: standardized Lp; σ: small-world properties.

2.3 节点属性分析

2.3.1 节点度中心性分析

       基于KLDs和JSDs方法计算节点DC得出了一致的结果。与HC组相比,mTBI组只有皮层厚度指标的节点DC在左侧额极横回、中央沟、外侧沟水平支和右侧枕极显著增强(P<0.05,FDR校正);在左侧楔叶、角回、海马旁回显著降低(P<0.05,FDR校正)。见表2图3

图3  两组间基于JSDs的节点DC差异性脑区图。mTBI组与HC组节点DC比较,蓝色小球表示mTBI组较HC组节点DC减低的脑区,蓝色小球体积越大,表示减低越明显;红色小球表示mTBI组较HC组节点DC增强的脑区,红色小球体积越大,表示增强越明显。
图4  两组间基于JSDs的Ne差异脑区图。mTBI组与HC组Ne比较,蓝色小球表示mTBI组较HC组Ne减低的脑区,蓝色小球体积越大,表示减低越明显;红色小球表示mTBI组较HC组Ne增强的脑区,红色小球体积越大,表示增强越明显。JSDs:Jensen-Shannon散度相似性;DC:度中心性;mTBI:轻度创伤性脑损伤;HC:健康对照;Ne:节点效率;tFP:额极横回;cs:中央沟;half:外侧沟水平支;ppo STG:颞上回极平面;CUN:楔叶;ANG:角回;PHG:海马旁回;PCUN:楔前叶;cas:胼胝体沟;OP:枕极;L:左;R:右。
Fig. 3  Node efficiency difference brain area map between two groups based on JSDs. Comparison of nodular degree centrality between mTBI and HC groups. The blue globules represent the brain regions with reduced node degree centrality in mTBI group compared to HC group. The larger the blue globules, the more obvious the reduction; the red globules represent brain regions with increased node degree centrality in mTBI group compared to HC group. The larger the red globules, the more obvious the enhancement.
Fig. 4  Brain area map of node efficiency differences between two groups based on JSDs. Comparison of nodular efficiency between mTBI and HC groups. The blue globules represent the brain regions with reduced node efficiency in mTBI group compared to the HC group. The larger the blue globules, the more obvious the reduction; The red globules represents the brain regions with increased node efficiency in the mTBI group compared to the HC group. The larger the red globules, the more obvious the enhancement. JSDs: Jensen-Shannon divergence-based similarity; DC: degree centrality; mTBI: mild traumatic brain injury; HC: healthy control; Ne: nodal efficiency; tFP: frontal pole transverse gyrus; cs: central sulcus; half: horizontal anterior segment of lateral fissure; ppo STG: planum polare of the superior temporal gyrus; CUN: cuneus; ANG: angular return; PHG: parahippocampal gyrus; PCUN: Precuneus; cas: callosal sulcus; OP: pole occipital; L: left; R: right.
表2  基于JSDs的HC组与mTBI组度中心性比较
Tab. 2  Comparison of nodular degree centrality between mTBI and HC groups based on JSDs

2.3.2 节点效率分析

       基于KLDs和JSDs方法计算Ne得出了一致的结果。与HC组相比,mTBI组只有皮层厚度指标的Ne在左侧额极横回、中央沟、外侧沟水平支和右侧楔前叶显著增强(P<0.05,FDR校正);在左侧颞上回极平面、角回、海马旁回和胼胝体沟显著减弱(P<0.05,FDR校正)。见表3图4

表3  基于JSDs的HC组与mTBI组节点效率比较
Tab. 3  Comparison of nodular efficiency between mTBI and HC groups based on JSDs

3 讨论

       本研究利用图论分析的方法探究了急性期mTBI患者个体形态学脑网络的全局及节点属性变化。结果表明:(1)虽然mTBI患者的形态学脑网络保持了小世界属性,但全局拓扑属性并无统计学意义;(2)在节点属性上,皮层厚度节点DC和Ne明显增高,主要集中在左眶额、额顶、顶叶和右枕极皮层;节点DC和Ne明显降低,主要集中在左颞缘、边缘和右枕叶皮层。这些结果为理解急性期mTBI的形态学脑网络改变提供了新的视角。

3.1 mTBI患者全局拓扑属性变化

       人类大脑网络通常被认为具备“小世界网络(small-worldness network)”属性,该属性能促进大脑对信息的分离和整合,为调节个体的认知水平提供一个强大的结构框架[12]。本研究中,mTBI组各形态学指标的形态学脑网络仍具有小世界属性,这与KUANG等[13]的研究一致,说明mTBI组在功能分化和信息整合之间仍保持相对平衡的状态。尽管在全局拓扑属性上,两组组间无统计学意义,但观察到mTBI组的全局属性Cp、Lp、γ、σ、Eloc轻度升高,λ、Eglob轻度减低的趋势。这些微弱的变化可能提示其网络特征向更弱的小世界属性转化的趋势,这与之前基于DTI的脑结构网络研究和基于fMRI的脑功能网络研究结论相一致[14, 15],此外,细微的差异也可能是由患者纳入标准的差异所致。然而在小儿急性mTBI中,mTBI组的Cp、Eloc明显减低[16, 17]。可能的原因是:大脑网络的全局拓扑属性与年龄密切相关,随着儿童脑发育趋于成熟,越来越多的短连接和长连接共同作用,使其分离和整合信息的效率逐渐提高[4, 18]。因此,成人在经历急性mTBI后,大脑网络也能在一定程度上保持稳态。

3.2 mTBI患者节点属性变化及其相应脑区

       节点DC和Ne是图论分析的两个关键指标,节点的DC越高,它在信息传输和集成中起到的作用越核心;Ne越高,节点在网络中进行信息交流的效率越高;二者共同反映了大脑对信息处理、决策制定和执行能力[19]

       本研究发现急性mTBI组的皮层厚度节点DC和Ne明显增高,主要集中在左侧额极横回、中央沟、外侧沟水平支和右侧枕极、楔前叶。既往研究表明,其中一些脑区与不良的预后有关,尤其是左侧额极横回和右侧楔前叶。额极横回作为前额叶皮层的一部分,在高级认知功能(如计划、解决问题、推理和情景记忆检索)中起着核心作用[20]。在经历突发事件时,前额叶是最易受到损伤的部位,因此,为了适应创伤后在行为和认知上的改变,大脑首先在前额叶对动态环境信息和高级认知功能采取一种自我调整[21, 22]。在解剖学上,楔前叶与前额叶同属于默认模式网络,楔前叶作为默认模式网络的关键节点,参与认知功能,例如执行功能、工作记忆和有意识的信息处理。研究表明,急性期mTBI从突显网络到楔前叶的有效连接增加与较差的神经认知评分呈负相关[23]。本研究中,发现左侧额极横回、右侧楔前叶等多个脑区的节点DC和Ne升高,其中左侧额极横回的异常变化与先前的研究不一致,推测急性期mTBI可能存在对认知功能受损的代偿机制,mTBI后的最初几天,受损的神经元需要增强连接性才能产生与创伤前相同的信号。

       与HC组相比,mTBI组的左侧楔叶、角回、海马旁回、颞上回极平面、角回、胼胝体沟的皮层厚度节点DC和Ne明显减低。其中,以左海马旁回的节点DC、Ne差异性最为显著。海马旁回是海马的主要皮质结构之一,位于海马结构的两侧,主要负责记忆及情绪调节[24, 25, 26]。该皮层连接性的减低可能反映了局部皮层信息整合的减少,即某些节点在信息处理中的作用下降。研究表明,随着岛叶皮层厚度的恢复,mTBI患者头痛症状与认知功能障碍能得到有效改善[27, 28];并且海马旁回脑血流量的减少与更严重的愤怒、焦虑和抑郁症状有关[29, 30, 31]。结合本研究,左侧海马旁回节点DC和Ne明显减低,表明该脑区在mTBI的病理生理机制中扮演重要角色。大量研究表明,轻度认知障碍、阿尔茨海默病和痴呆的情绪、认知障碍与海马旁回息息相关,海马旁回是最早受损的区域之一[32, 33, 34]。本研究中mTBI组的左海马旁回节点DC和Ne明显减低,与先前的研究一致[35]。这对于早期发现mTBI患者潜在的情绪、认知能力下降具有重要意义。

       此外,本研究发现左侧大脑半球发生节点属性改变的脑区明显比右侧多,这与之前的研究结果类似,VEKSLER等[36]发现mTBI患者白质完整性的降低主要集中在左侧下额枕束、上纵束和下纵束;DENNIS等[37]也发现TBI患者中扣带回的侧向化改变,扩散张量成像(diffusion tensor imaging, DTI)中左侧扣带回分数各向异性明显增强。原因可能是,灰质不对称可能与纤维束不对称共同促进大脑的功能侧向化,由于患者均是右利手,因此优势大脑半球为左侧大脑半球,左侧大脑半球对大脑微环境变化的敏感性更高,更容易受到脑外伤的影响。

3.3 局限性及展望

       本研究具有一定的局限性。首先,本研究样本量较小,有研究表明,基于个体形态学脑网络的构建随受试者数量的增加而逐渐趋于稳定(数量>70人),将来我们将扩大样本量来验证研究结果的可重复性[3]。其次,本研究采用结构神经图像形态指标的概率分布方法构建个体形态学脑网络。因此,形态学脑网络的构建取决于脑图谱和相似性评估方法的选择,在后续的研究中,可以采用其他一种或几种相似性指标评估脑区连接的相似性,验证方法的可行性。最后,在本研究中,我们使用二进制网络而不是加权网络模型表征形态学脑网络。加权网络模型可以提供更多形态学脑网络相关拓扑组织的信息,能更敏感地捕捉大脑在发生突发事件(疾病、发育和衰老)时的形态脑网络变化[38]

4 结论

       综上所述,mTBI患者表现出个体形态学脑网络的低节点DC和低Ne,其局部信息整合及处理能力下降,且主要影响记忆及情绪处理相关的皮层。这些结果表明异常形态学脑网络模式(低节点DC和低Ne)对mTBI神经病理学的关键贡献。节点DC及Ne可能成为潜在的mTBI影像生物标志物,需要对其进行更全面地描述,以便充分了解mTBI发病机制和行为症状。

[1]
ZHENG X Y, YI Q, XU X J, et al. Trends and external causes of traumatic brain injury and spinal cord injury mortality in south China, 2014-2018: an ecological study[J/OL]. BMC Public Health, 2021, 21(1): 2225 [2023-08-23]. https://pubmed.ncbi.nlm.nih.gov/34876065/. DOI: 10.1186/s12889-021-12225-2.
[2]
WONG J K Y, CHURCHILL N W, GRAHAM S J, et al. Altered connectivity of default mode and executive control networks among female patients with persistent post-concussion symptoms[J]. Brain Inj, 2023, 37(2): 147-158. DOI: 10.1080/02699052.2022.2163290.
[3]
BRENNAN D, DELANEY C, FARRELL M, et al. Polypathology-associated neurodegeneration after remote head injury[J]. Clin Neuropathol, 2023, 42(6): 201-211. DOI: 10.5414/NP301576.
[4]
LI Y, WANG N, WANG H, et al. Surface-based single-subject morphological brain networks: Effects of morphological index, brain parcellation and similarity measure, sample size-varying stability and test-retest reliability[J/OL]. Neuroimage, 2021, 235: 118018 [2023-08-23]. https://pubmed.ncbi.nlm.nih.gov/33794358/. DOI: 10.1016/j.neuroimage.2021.118018.
[5]
HE C, CORTES J M, KANG X, et al. Individual-based morphological brain network organization and its association with autistic symptoms in young children with autism spectrum disorder[J]. Hum Brain Mapp, 2021, 42(10): 3282-3294. DOI: 10.1002/hbm.25434.
[6]
ZHANG W, DUAN Y, QI L, et al. Distinguishing patients with MRI-negative temporal lobe epilepsy from normal controls based on individual morphological brain network[J]. Brain Topogr, 2023, 36(4): 554-565. DOI: 10.1007/s10548-023-00962-z.
[7]
HOMAN P, ARGYELAN M, DEROSSE P, et al. Structural similarity networks predict clinical outcome in early-phase psychosis[J]. Neuropsychopharmacology, 2019, 44(5): 915-922. DOI: 10.1038/s41386-019-0322-y.
[8]
ZHENG W, EILAMSTOCK T, WU T, et al. Multi-feature based network revealing the structural abnormalities in autism spectrum disorder[J]. IEEE Trans Affect Comput, 2019, 12(3): 732-742. DOI: 10.1109/TAFFC.2018.2890597.
[9]
Lin N, Wu S, Ji S. A morphologically individualized deep learning brain injury model[J]. J Neurotrauma, 2023, 40(19-20): 2233-2247. DOI: 10.1089/neu.2022.0413.
[10]
KONG XZ, WANG X, HUANG L, et al. Measuring individual morphological relationship of cortical regions[J]. J Neurosci Methods, 2014, 237: 103-107. DOI: 10.1016/j.jneumeth.2014.09.003.
[11]
WANG H, JIN X, ZHANG Y, et al. Single-subject morphological brain networks: connectivity mapping, topological characterization and test-retest reliability[J/OL]. Brain Behav, 2016, 6(4): e00448 [2023-08-23]. https://pubmed.ncbi.nlm.nih.gov/27088054/. DOI: 10.1002/brb3.448.
[12]
FORNITO A, ZALESKY A, BREAKSPEAR M. Graph analysis of the human connectome: promise, progress, and pitfalls[J]. Neuroimage, 2013, 80: 426-444. DOI: 10.1016/j.neuroimage.2013.04.087.
[13]
KUANG H M, CHEN Y, HUANG J L, et al. Acute changes in the resting brain networks in concussion patients: Small-world topology perspective[J/OL]. J Integr Neurosci, 2024, 23(1): 12 [2023-08-23]. https://pubmed.ncbi.nlm.nih.gov/38287842/. DOI: 10.31083/j.jin2301012.
[14]
WANG S, GAN S, YANG X, et al. Decoupling of structural and functional connectivity in hubs and cognitive impairment after mild traumatic brain injury[J]. Brain Connect, 2021, 11(9): 745-758. DOI: 10.1089/brain.2020.0852.
[15]
VIRJI-BABUL N, HILDERMAN C G, MAKAN N, et al. Changes in functional brain networks following sports-related concussion in adolescents[J]. J Neurotrauma, 2014, 31(23): 1914-1919. DOI: 10.1089/neu.2014.3450.
[16]
WARE A L, YEATES K O, GEERAERT B, et al. Structural connectome differences in pediatric mild traumatic brain and orthopedic injury[J]. Hum Brain Mapp, 2022, 43(3): 1032-1046. DOI: 10.1002/hbm.25705.
[17]
WARE A L, ONICAS A I, ABDEEN N, et al. Altered longitudinal structural connectome in paediatric mild traumatic brain injury: an advancing concussion assessment in paediatrics study[J/OL]. Brain Commun, 2023, 5(3): fcad173 [2023-08-23]. https://pubmed.ncbi.nlm.nih.gov/37324241/. DOI: 10.1093/braincomms/fcad173.
[18]
IMMS P, CLEMENTE A, COOK M, et al. The structural connectome in traumatic brain injury: A meta-analysis of graph metrics[J]. Neurosci Biobehav Rev, 2019, 99: 128-137. DOI: 10.1016/j.neubiorev.2019.01.002.
[19]
HENRY T R, DUFFY K A, RUDOLPH M D, et al. Bridging global and local topology in whole-brain networks using the network statistic jackknife[J]. Netw Neurosci, 2020, 4(1): 70-88. DOI: 10.1162/netn_a_00109.
[20]
BRAVER T S, BONGIOLATTI S R. The role of frontopolar cortex in subgoal processing during working memory[J]. Neuroimage, 2002, 15(3): 523-536. DOI: 10.1006/nimg.
[21]
HUANG W, HU W, ZHANG P, et al. Early changes in the white matter microstructure and connectome underlie cognitive deficit and depression symptoms after mild traumatic brain injury[J/OL]. Front Neurol, 2022, 13: 880902 [2023-08-23]. https://pubmed.ncbi.nlm.nih.gov/35847204/. DOI: 10.3389/fneur.2022.880902.
[22]
CATANI M. The anatomy of the human frontal lobe[J]. Handb Clin Neurol, 2019, 163: 95-122. DOI: 10.1016/B978-0-12-804281-6.00006-9.
[23]
LU L, LI F, MA Y, et al. Functional connectivity disruption of the substantia nigra associated with cognitive impairment in acute mild traumatic brain injury[J]. Eur J Radiol, 2019, 114: 69-75. DOI: 10.1016/j.ejrad.2019.03.002.
[24]
FORTIER-LEBEL O, JOBIN B, LÉCUYER-GIGUÈRE F, et al. Verbal episodic memory alterations and hippocampal atrophy in acute mild traumatic brain injury[J]. J Neurotrauma. 2021, 38(11): 1506-1514. DOI: 10.1089/neu.2020.7475.
[25]
熊峰, 李天惠, 潘怡臻, 等. 动脉自旋标记磁共振评估轻度脑外伤患者的脑血流变化[J]. 中南大学学报(医学版), 2022, 47(8): 1016-1024. DOI: 10.11817/j.issn.1672-7347.2022.210754.
XIONG F, LI T H, PAN Y Z, et al. Arterial spin labeling magnetic resonance evaluates changes of cerebral blood flow in patients with mild traumatic brain injury[J]. Journal of Central South University (Medical Science), 2022, 47(8): 1016-1024. DOI: 10.11817/j.issn.1672-7347.2022.210754.
[26]
KUMARI M, ARORA P, SHARMA P, et al. Acute metabolic alterations in the hippocampus are associated with decreased acetylation after blast induced TBI[J/OL]. Metabolomics. 2023, 19(1): 5 [2023-08-23]. https://pubmed.ncbi.nlm.nih.gov/36635559/. DOI: 10.1007/s11306-022-01970-z.
[27]
XU H, XU C, GU P, et al. Neuroanatomical restoration of salience network links reduced headache impact to cognitive function improvement in mild traumatic brain injury with posttraumatic headache[J/OL]. J Headache Pain, 2023, 24(1): 43 [2023-08-23]. https://pubmed.ncbi.nlm.nih.gov/37081382/. DOI: 10.1186/s10194-023-01579-0.
[28]
MAYER A R, MEIER T B, DODD A B, et al. Prospective study of gray matter atrophy following pediatric mild traumatic brain injury[J/OL]. Neurology, 2023, 100(5): e516-e527 [2023-08-23]. https://pubmed.ncbi.nlm.nih.gov/36522161/. DOI: 10.1212/WNL.0000000000201470.
[29]
LI F, LU L, SHANG S, et al. Cerebral blood flow and its connectivity deficits in mild traumatic brain injury at the acute stage[J/OL]. Neural Plast, 2020, 2020: 2174371 [2023-08-23]. https://pubmed.ncbi.nlm.nih.gov/32684919/. DOI: 10.1155/2020/2174371.
[30]
MA J, HONG G, HA E, et al. Hippocampal cerebral blood flow increased following low-pressure hyperbaric oxygenation in firefighters with mild traumatic brain injury and emotional distress[J]. Neurol Sci, 2021, 42(10): 4131-4138. DOI: 10.1007/s10072-021-05094-5.
[31]
PAPADAKI E, KAVROULAKIS E, MANOLITSI K, et al. Cerebral perfusion disturbances in chronic mild traumatic brain injury correlate with psychoemotional outcomes[J]. Brain Imaging Behav, 2021, 15(3): 1438-1449. DOI: 10.1007/s11682-020-00343-1.
[32]
HAYES J P, LOGUE M W, SADEH N, et al. Mild traumatic brain injury is associated with reduced cortical thickness in those at risk for Alzheimer's disease[J]. Brain, 2017, 140(3): 813-825. DOI: 10.1093/brain/aww344.
[33]
CHENG C P, CHENG S T, Tam C W, et al. Relationship between cortical thickness and neuropsychological performance in normal older adults and those with mild cognitive impairment[J]. Aging Dis, 2018, 9(6): 1020-1030. DOI: 10.14336/AD.2018.0125.
[34]
LI M J, HUANG S H, HUANG C X, et al. Morphometric changes in the cortex following acute mild traumatic brain injury[J]. Neural Regen Res, 2022, 17(3): 587-593. DOI: 10.4103/1673-5374.320995.
[35]
LIU Y, LI F, SHANG S, et al. Functional-structural large-scale brain networks are correlated with neurocognitive impairment in acute mild traumatic brain injury[J]. Quant Imaging Med Surg, 2023, 13(2): 631-644. DOI: 10.21037/qims-22-450.
[36]
VEKSLER R, VAZANA U, SERLIN Y, et al. Slow blood-to-brain transport underlies enduring barrier dysfunction in American football players[J]. Brain, 2020, 143(6): 1826-1842. DOI: 10.1093/brain/awaa140.
[37]
DENNIS E L, NEWSOME M R, LINDSEY H M, et al. Altered lateralization of the cingulum in deployment-related traumatic brain injury: An ENIGMA military-relevant brain injury study[J]. Hum Brain Mapp, 2023, 44(5): 1888-1900. DOI: 10.1002/hbm.26179.
[38]
MIJALKOV M, VERÉB D, JAMIALAHMADI O, et al. Sex differences in multilayer functional network topology over the course of aging in 37543 UK Biobank participants[J]. Netw Neurosci, 2023, 7(1): 351-376. DOI: 10.1162/netn_a_00286.

上一篇 多序列MRI影像组学在脑膜瘤鉴别诊断中的价值
下一篇 3 T MRI定量测量眼眶结构在Graves眼病分期中的诊断价值
  
诚聘英才 | 广告合作 | 免责声明 | 版权声明
联系电话:010-67113815
京ICP备19028836号-2