复发缓解型多发性硬化全脑白质磁共振扩散张量成像研究

发布时间：2018-05-12 14:01

本文选题：复发缓解型 + 多发性硬化　；参考：《南方医科大学》2014年硕士论文

【摘要】：[目的] 利用3.0T Philips磁共振扩散张量成像技术(DTI)并采用基于纤维束示踪的空间统计(TBSS)的方法观察复发缓解型多发性硬化患者全脑白质微结构改变,评价DTI两个参数(FA、 MD/ADC)值在多发性硬化患者脑白质微结构及视觉通路损伤中的应用价值,探索全脑白质、视觉通路纤维束损害的特点以及可能的发病机制,从而为多发性硬化患者的临床诊断提供客观的影像学依据。 [材料及方法] 1.研究对象本研究搜集2012年8月至2014年3月在广州南方医科大学南方医院影像中心接受MRI检查的复发缓解型多发性硬化(Relapsing-remitting Multiple Sclerosis, RRMS)患者共21例,男6例,女15例,年龄27-61岁,平均45.48±12.87岁,受教育程度6-16年,平均12.85±3.55年；病程(首次发病到最近一次发病时间)0.04-15.5年,平均7.09±5.46年；复发缓解次数2-7次,平均3.71±1.85年。所有RRMS患者符合2010年McDonald的MS诊断标准,磁共振检查前两个月内未使用过糖皮质激素、免疫抑制剂等药物。正常对照组21例(性别、年龄、文化程度与RRMS组差异无统计学意义),男9例,女12例,年龄26-60岁,平均年龄46.71±11.68岁,受教育程度6-16年,平均13.35±2.85年。无任何临床不适症状,无颅内疾病病史及外伤史,无认知功能障碍,行常规MRI检查头颅及视觉通路未见异常信号改变。所有纳入的研究对象对本实验均知情同意。 2.扫描设备及序列磁共振数据采集使用飞利浦3.0T磁共振扫描仪(Philips, Achieva,the Netherlands)并采用标准8通道头颅线圈接收核磁共振信号。DTI数据采集之前均进行全脑轴位T1WI、T2WI、 FLAIR扫描。头颅常规平扫参数如下：轴位TSET2WI (TR/TE=3000/80ms),轴位T1WI (TR/TE=2000/120ms),轴位FLAIR (TR/TE=9000/120ms);翻转角(flip angle)=90°;矩阵(matrix size)=460×268;视野(FOV)=230mm×230mm;激励次数(NEX)=1；层厚(thickness)=6.Omm;层数(slice)=20;层间距(slice gap)=0。头颅常规DTI扫描参数：采集使用单次激发自旋回波平面回波序列,SPIR压脂扫描,扫描以大脑前-后联合连线平面为参考平面。TR/TE=9000/85ms, FOV=230mm×230mm,矩阵=92x108,b=0,1000s/mm2,层厚=2mm,层间距0,NEX=2,扩散方向32个,共扫描60层,扫描时间9min27s。视觉通路DTI扫描参数同常规DTI扫描参数,扫描以视交叉-后联合(CH-PC)连线平面为参考平面。扫描时受试者全程保持头部固定,双眼闭合。3.DTI数据处理 3.1全脑白质数据处理首先采用MRIcron(http://www.mccauslandcenter.sc.edu/mricro/mricron/install)将每个被试者的1组B0图像与相应的32组扩散加权二维图像的DICOM数据格式转化为四维NIFTI数据格式,其中包含了三维影像,并代表不同的扩散权重梯度场强方向。 DTI数据分析采用FSL (http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FSL5.0.6)软件包。具体处理步骤包括：①使用FSL软件中的Eddy correct函式对转换好的DTI数据进行头动及涡流校正。②利用FSL软件中的BET函式,以每个被试者的B0图像作为依据产生各自的脑Mask。③通过FDT中的DTIFIT函式,将梯度磁场方向、磁场强度的数值以及四维NIFTI格式图像,连同各个被试的Mask输入DTIFIT函式进行扩散张量计算。输出的结果将包含部分各向异性(fractional anisotropy, FA)、平均扩散率(mean diffusivity、及相对应的三个特征向量(V1、V2、V3)。然后按TBSS四部标准流程进行分析,包括tbss_1_preproc、 tbss_2_reg-T、 tbss_3_postreg-S、tbss_4_prestat，利用Randomize函式对两组的FA值进行基于纤维素示踪的全脑非参数统计比较,以P0.05作为统计检验水准,加入无阈值簇增强(threshold-free cluster enhancement，TFCE)选项以校正多重比较中的I类错误。再将所有受试者各自的FA图非线性配准到MD图上,生成配准后的参数图,并将之投射到平均FA纤维骨架上,进行基于纤维素示踪的全脑非参数统计比较。 3.2双侧视神经评价将扫描的原始DTI图像传至Philips后处理工作站。使用Diffusion软件获得各向异性分数(FA)图、表观弥散系数(ADC)图及彩色编码图,结合T2WI序列,选取双侧视神经球后段,在FA图、ADC图上画感兴趣区(region of interest,ROI),大小为5.5mm2,测量双侧视神经球后段FA值及ADC值,测量三次取平均值。 4.统计学分析所得数据均以均数±标准差(x±s)来表示,应用SPSS13.0软件两独立样本t检验的统计方法比较复发缓解型多发性硬化组与正常对照组之间年龄、受教育程度的差异,卡方检验评价两被试组之间性别差异。全脑白质纤维骨架平均FA值、MD值图组间基于体素的比较,采用Randomize函式进行非参数统计阈值分析的统计方法,置换测试次数5000,采用以簇为基础的校正方法进行多重比较校正。P0.05且簇50个体素(体素大小：1×1×1mm3)认为有统计学意义。P0.01认为差异有显著性统计学意义。采用偏相关分析方法,以患者年龄、复发缓解次数为协变量,比较病程与全脑有显著性差异脑区的FA值、MD值的相关性；以年龄、病程为协变量,比较复发缓解次数与全脑有显著性差异脑区的FA值、MD值的相关性,P0.05认为有统计学意义。 [结果] 1.脑白质分析结果 RRMS患者全脑双侧额叶、颞叶、顶叶及枕叶脑白质区域纤维素呈广泛性FA值减低,MD值升高,差异有统计学意义(P0.05)；左侧上纵束、双侧下纵束、左侧上枕额束、双侧下额枕束、胼胝体膝部、压部、右侧扣带束、左侧钩束、右侧内囊前肢、双侧外囊FA值明显减低,与对照组比较差异有显著统计学意义(P0.01)；左侧下纵束、双侧下枕额束、胼胝体膝部、压部、右侧扣带束、左侧钩束、右侧内囊前肢、双侧外囊FA值显著减低(P0.001)。左侧上纵束、双侧下纵束、双侧上枕额束、双侧下额枕束、胼胝体膝部、压部、双侧扣带束、双侧钩束、左侧内囊前肢、左侧内囊后肢、双侧外囊MD值明显升高,差异有显著统计学意义(P0.01)。双侧下纵束、双侧上枕额束、胼胝体膝部、压部、左侧扣带束、双侧钩束、左侧内囊后肢MD值较对照组显著升高(P0.001)。RRMS组未发现FA值明显增高、MD值明显减低的纤维束。偏相关性分析：以年龄及复发缓解次数为协变量,胼胝体膝部(r=-0.539,P=0.017)、右内囊前肢(r=-0.524,P=0.016)的FA值与病程(首次发病至最近一次发病)呈负相关；胼胝体膝部(r=0.514,P=0.024)、左上纵束(r=0.479,P=0.038)的MD值与病程呈正相关(P0.05)。其余组间有差异的脑白质区FA值、MD值与病程之间的相关性无统计学意义(P0.05)。以年龄及病程为协变量,各组间有差异的脑白质区FA值、MD值与复发缓解次数之间的相关性无统计学意义(P0.05)。 2.视觉通路分析结果： RRMS组双侧视神经、视束及视辐射FA值较正常对照组减低,差异有统计学意义(P0.05)；左侧视神经、双侧视束及视辐射FA值较正常对照组减低,差异有显著性统计学意义(P0.01)。双侧视束及双侧视辐射MD值较正常对照组升高,差异有显著统计学意义(P0.01),双侧视神经MD值两组间比较差异无统计学意义(P0.05)。偏相关性分析：以年龄及复发缓解次数为协变量,RRMS患者右侧视束(r=-0.545,P=0.016)的FA值与病程呈负相关。其余组间FA值、MD值与病程相关性无统计学意义(P0.05)。以年龄及病程为协变量,RRMS患者左侧视辐射(r=-0.531,P=0.019)的FA值与复发缓解次数呈负相关(P0.05)。其余组间FA值、MD值与复发缓解次数之间的相关性无统计学意义(P0.05)。 [结论] l.RRMS患者双侧额叶、颞叶、顶叶及枕叶脑白质区存在广泛的脑白质损伤,其中以左侧上纵束、双侧下纵束、双侧上枕额束、双侧下额枕束、胼胝体膝部、压部、双侧扣带束、双侧钩束、双侧内囊前肢、左侧内囊后肢及双侧外囊损伤更明显。 2.在RRMS患者白质损伤区域中,胼胝体膝部、右侧内囊前肢的FA值与病程呈负相关：胼胝体膝部、左侧上纵束的MD值与病程呈正相关。 3.RRMS患者视觉通路中双侧视神经、双侧视束及双侧视辐射存在损伤。 4.RRMS患者右侧视束的FA值与病程呈负相关,左侧视辐射的FA值与复发缓解次数呈负相关。 5.DTI在复发多发性硬化患者中有很高的诊断价值,TBSS后处理方法能够能在活体细胞水平对脑白质异常的区域进行更准确的定位,定量评估脑白质的完整性及损伤,可监测全脑的白质及视觉通路的潜在性损伤,为临床提供客观定量的影像学依据。
[Abstract]:[Objective]
3.0T Philips magnetic resonance diffusion tensor imaging (DTI) and spatial statistics (TBSS) based on fiber bundle tracer (TBSS) were used to observe the changes in the white mass structure of the whole brain of the patients with relapsed remission multiple sclerosis, and evaluate the application of the two parameters of DTI (FA, MD/ADC) in the white mass structure and visual pathway damage of the patients with multiple hardened brain. To explore the characteristics of the white matter in the whole brain, the characteristics of the damage of the fiber bundle of the visual pathway and the possible pathogenesis, so as to provide an objective imaging basis for the clinical diagnosis of the patients with multiple sclerosis.
[materials and methods]
1. research objects
This study collected 21 cases of recurrent remission multiple sclerosis (Relapsing-remitting Multiple Sclerosis, RRMS) in the imaging center of Southern Hospital of Southern Medical University, Guangzhou, from August 2012 to March 2014. There were 6 males and 15 females, 27-61 years old, with an average of 45.48 + 12.87 years old. The average education level was 6-16 years, and the average was 12.85 + 3.55 years. The course of disease (first onset to the most recent onset time) was 0.04-15.5 years, average 7.09 + 5.46 years, relapse remission times 2-7 times, averaging 3.71 + 1.85 years. All RRMS patients were in line with the MS diagnostic criteria of McDonald in 2010, and no glucocorticoids and immunosuppressive drugs were used within two months before magnetic resonance examination.
There were 21 cases of normal control group (sex, age, education level and RRMS group), 9 men, 12 women, 26-60 years old, and the average age was 46.71 + 11.68 years old. The average age was 46.71 + 11.68 years old. The average age of education was 6-16 years, and the average was 13.35 + 2.85 years. There was no clinical discomfort symptoms, no history of intracranial disease and trauma, no cognitive dysfunction, routine MRI examination of skull and No abnormal signal changes were found in the visual pathway. All participants were informed and agreed with the experiment.
2. scanning equipment and sequence
Magnetic resonance data acquisition using the PHILPS 3.0T magnetic resonance scanner (Philips, Achieva, the Netherlands) and using a standard 8 channel head coil to receive NMR signal.DTI data collected before the.DTI data acquisition of all brain axis T1WI, T2WI, FLAIR scan. 2000/120ms), axis FLAIR (TR/TE=9000/120ms), turning angle (flip angle) =90 degree; matrix (matrix size) =460 x 268; visual field (FOV) =230mm x 230mm;
Routine DTI scan parameters of head: acquisition and use of single excited spin echo plane echo sequence, SPIR compression scan, scanning with front and back joint plane of brain as reference plane.TR/TE=9000/85ms, FOV=230mm x 230mm, matrix =92x108, b=01000s/mm2, layer thickness =2mm, interval 0, NEX=2, and diffusion direction 32, scanning 60 layers, scanning time 9min27 S.
The DTI scanning parameters of the visual pathway and the conventional DTI scanning parameters were scanned with the optic cross and post Union (CH-PC) line plane as the reference plane. The subjects kept the head fixed throughout the scan and the.3.DTI data processing in the eyes closed.
Data processing of 3.1 white matter in the whole brain
First, MRIcron (http://www.mccauslandcenter.sc.edu/mricro/mricron/install) is used to transform 1 groups of B0 images of each participant and the corresponding DICOM data format of 32 groups of diffusion-weighted images into four dimensional NIFTI data format, which contains three dimensional images and represents the different direction of the gradient field strength of the different diffusion weight.
The DTI data analysis uses the FSL (http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FSL5.0.6) software package. The specific processing steps include: (1) using the Eddy correct function of the FSL software to carry out the head motion and eddy current correction of the converted DTI data. (2) using the BET function in FSL software to produce their respective brain Ma on the basis of the B0 images of each person. Sk. (3) through the DTIFIT function in FDT, the gradient magnetic field direction, the magnetic field intensity value and the four dimensional NIFTI format image are calculated with the Mask input DTIFIT function of each test. The output results will include the partial anisotropy (fractional anisotropy, FA), the average diffusivity (mean diffusivity, and the corresponding three). Feature vectors (V1, V2, V3). Then analyzed by four standard TBSS processes, including tbss_1_preproc, tbss_2_reg-T, tbss_3_postreg-S, tbss_4_prestat, using the Randomize function to carry out the non parametric statistical comparison of the two groups of FA values based on cellulose tracers, taking P0.05 as the statistical test level, adding no threshold cluster enhancement (threshold-). The free cluster enhancement, TFCE) option is used to correct the I class errors in multiple comparisons. Then, all the FA diagrams of all the subjects are registered on the MD graph, and the registration parameters are generated and projected onto the average FA fiber skeleton for non parametric statistical comparison of the whole brain based on cellulose tracers.
3.2 bilateral optic nerve evaluation
The scanned original DTI images were passed to the Philips post processing workstation. The Diffusion software was used to obtain the anisotropic fraction (FA), the apparent dispersion coefficient (ADC) and color coding, and the posterior segment of the bilateral optic nerve ball was selected with the T2WI sequence, and the region of interest (region of interest, ROI) was drawn on the FA map and ADC diagram, and the size was measured, and the bilateral optic gods were measured. The average FA value and ADC value of the posterior segment were measured three times.
4. statistical analysis
The data were all represented by mean mean standard deviation (x + s), and the statistical method of SPSS13.0 software two independent sample t test was used to compare the age between the relapsing remission multiple sclerosis group and the normal control group and the difference in education degree. The chi square test evaluated the gender difference between the two groups. The average FA value of the whole brain white matter fiber skeleton and the MD value group Based on the comparison of voxel, the statistical method of non parametric statistical threshold analysis was carried out by Randomize function. The number of replacement tests was 5000. The cluster based correction method was used for multiple comparison correction.P0.05 and cluster 50 elements (voxel size: 1 * 1 x 1mm3) considered statistically significant.P0.01. Using the method of partial correlation analysis, the patients' age and relapse remission times were co variables, the FA value and the correlation of MD value were compared between the course of the disease and the whole brain, and the age and the course of disease were co variables. The correlation of the FA value and the MD value of the relapse remission times with the whole brain was compared, and the P0.05 was statistically significant.
[results]
1. brain white matter analysis results
RRMS patients with bilateral frontal lobes, temporal lobes, parietal lobe and occipital lobes had extensive FA values and increased MD values, with significant differences (P0.05); left superior longitudinal fascicles, bilateral inferior longitudinal fascicles, bilateral inferior occipital fasciculus, the genu of the corpus callosum, the right buckle band, left hook bundle, right internal capsule forelimb, bilateral outer capsule, F. The A value was significantly lower than that in the control group (P0.01); the left inferior longitudinal fasciculus, bilateral inferior occipital fasciculus, the genu of the corpus callosum, the right cingulate bundle, the left hook bundle, the right inner capsule front, and the bilateral outer capsule FA value decreased significantly (P0.001). The left superior longitudinal fasciculus, bilateral inferior longitudinal fascicles, bilateral occipital fasciculus, corpus callosum Knee, pressure, bilateral buckle, bilateral hook, left internal capsule forelimb, left internal capsule hind limbs, bilateral outer capsule MD value increased significantly, the difference was significant (P0.01). Bilateral lower longitudinal fasciculus, bilateral upper occipital fasciculus, corpus callosum, pressure part, left cingulate bundle, double hook bundle, left internal capsule posterior limb MD value was significantly higher than the control group (P0.001).RRMS group (P0.001) group.RRMS No significant increase in FA value and significant decrease in MD value were observed.
Partial correlation analysis: the FA value of the corpus callosum (r=-0.539, P=0.017) and the anterior limb of the right internal capsule (r=-0.524, P=0.016) was negatively correlated with the course of disease (the first onset of the first onset), the corpus callosum (r=0.514, P=0.024), the MD value of the upper left longitudinal bundle (r=0.479, P=0.038) in the corpus callosum was positively correlated with the course of the disease (P0.05). There was no significant correlation between the FA value of the white matter area between the other groups and the correlation between the MD value and the course of disease (P0.05). With age and course of disease as covariate, there was a difference in the FA value of the white matter area between each group, and the correlation between the MD value and the remission times was not statistically significant (P0.05).
2. visual pathway analysis results:
The FA value of bilateral optic nerve, optic tract and optic radiation in group RRMS was lower than that of the normal control group (P0.05). The left optic nerve, bilateral optic tract and visual radiation FA were lower than those of the normal control group, the difference was statistically significant (P0.01). The MD value of bilateral optic tract and bilateral optic radiation was higher than that of the normal control group, the difference was statistically significant There was no significant difference in P0.01 (MD) between the two groups (P0.05).
Partial correlation analysis: the FA value of the right visual tract (r=-0.545, P=0.016) of RRMS patients was negatively related to the course of disease with age and relapse remission times. There was no statistical significance between the FA value of the other groups and the correlation between the MD value and the course of disease (P0.05). The FA value of the left visual radiation (r=-0.531, P=0.019) in RRMS patients and the relapse remission of the RRMS patients were associated with the age and course of disease. There was a negative correlation between the number of times (P0.05). There was no statistically significant correlation between the FA value, MD value and relapse remission frequency among the rest groups (P0.05).
[Conclusion]
There were extensive brain white matter injuries in the frontal lobe, temporal lobe, parietal lobe and occipital lobe of l.RRMS patients, including the left superior longitudinal fasciculus, bilateral inferior longitudinal fasciculus, bilateral occipital frontal fasciculus, bilateral inferior occipital fasciculus, the corpus callosum, bilateral buckle band, bilateral hook tract, bilateral internal capsule forelimb, left internal capsule hind limbs and bilateral outer capsule.
2. in the area of white matter injury in RRMS patients, the FA value of the corpus callosum and the right side of the right internal capsule was negatively correlated with the course of the disease: the MD value of the corpus callosum and the left superior longitudinal bundle was positively correlated with the course of disease.
The bilateral optic nerve, bilateral optic tract and bilateral optic radiation were damaged in the visual pathway of 3.RRMS patients.
The FA value of the right optic tract in 4.RRMS patients is negatively correlated with the course of disease. The FA value of the left visual radiation is negatively correlated with the number of remission times.
5.DTI has a high diagnostic value in patients with recurrent multiple sclerosis. TBSS post-processing can be able to accurately locate the area of abnormal white matter in the living cell level, evaluate the integrity and damage of the brain white matter, monitor the potential damage of the white matter and visual pathway in the whole brain, and provide an objective quantitative shadow for the clinical. As a basis for learning.

【学位授予单位】：南方医科大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：R744.51;R445.2

【参考文献】