【作者】 于洋；

【导师】 祝一虹；

【作者基本信息】 浙江大学 ， 精神病与精神卫生学， 2015， 硕士

【摘要】 研究目的：伴有中央颞叶棘波的儿童良性癫痫(Benign epilepsy of childhood with centrotemporal spikes, BECTS)又名为rolandic癫痫(rolandic epilepsy).以往有研究发现BECTS患者在放电状态下神经活动异常。但是,有关无放电状态下BECTS患者的神经活动是否正常还不清楚。此外,BECTS患者的异常神经活动是否可以应用于区分个体水平无放电BECTS患者和正常对照也亟待探明。研究方法：共收集43名BECTS患者的同步脑电-功能磁共振数据和28名年龄和性别匹配的正常对照的静息态功能磁共振数据。利用同步采集的脑电图判断收集数据过程中患者是否放电,进而将BECTS患者分成两组：放电组(20人)和无放电组(23人)。利用低频振幅(Amplitude of low Frequency Fluctuation, ALFF)方法分别计算放电组、无放电组和正常对照组共3组被试的自发脑活动,用单因素方差分析统计方法来探究3组人脑自发活动的差异。再将放电组的BECTS患者按照放电侧别不同分为左侧放电组和右侧放电组,利用单因素方差分析统计方法来探究左侧放电组、右侧放电组和正常对照组组间的认知功能和自发脑活动是否存在差异。此外,采用多维模式分析方法区分无放电的BECTS患者与正常对照。研究结果：发现3组被试在右侧中央后回,右侧中央盖区,右侧丘脑,左侧顶上回,左侧额中回和左侧直回6个脑区的脑自发活动存在差异。放电组患者右侧丘脑、中央盖区和中央后回的低频振幅显著高于对照组,无放电组患者左侧额中回、右侧中央盖区和中央后回的低频振幅也与对照组有显著差异。放电组在右侧丘脑的低频振幅与同步记录的脑电放电次数成正相关关系(r=0.6088,p=0.0073)。此外,放电组和病人组在左侧顶上回的低频振幅与操作智商成负相关关系(r=-0.4822,P=0.05；r=-0.3299,P=0.0376),在无放电组无统计学相关关系(r=-0.2111,P=0.3335)。不同放电侧别的BECTS患者的总智商、言语智商和操作智商无显著差异。左侧放电BECTS患者和右侧放电BECTS患者的低频振幅与正常对照组相比在右侧中央盖区和右侧缘上回均有差异。利用多维模式分析(MVPA)方法可以将无放电的BECTS患者与正常对照区分开来(准确率为92.16%,p<0.001)。研究结论：不论有、无放电的BECTS患者在右侧中央盖区和中央后回的自发神经活动都与正常对照有显著差异。右侧丘脑可能与BECTS放电有关。利用低频振幅这个指标,并通过用多维模式分类方法能够从正常对照中区分开无放电BECTS患者,并为临床提供诊断帮助。更多还原

【Abstract】 Objective:The patients with rolandic epilepsy are usually associated with neural circuit dysfunction, particularly during the transient active state characterized by interictal epileptiform discharges (IED). Little is known, however, about neural circuits abnormalities in rolandic epilepsy without interictal epileptiform discharges, and whether such abnormal neural circuits would be used to differentiate rolandic epilepsy individuals without interictal epileptiform discharges from the healthy control.Methods:We collected resting-state functional magnetic resonance imaging (fMRI) and simultaneous electroencephalography from children with rolandic epilepsy (n=43) and healthy age-matched controls (n=28). We used simultaneous EEG to group rolandic epilepsy patients with IED (IED group, n=20) and without IED (non-IED group, n=23). Intrinsic brain activity was calculated among three groups using amplitude of low frequency fluctuation (ALFF). Then we made one-way ANOVA for the ALFF of the three groups to find group differences of intrinsic brain activity. Moreover, we grouped patients with left IED and right IED and compared the IQs and amplitude of low frequency fluctuation (ALFF) among patients with left IED and right IED and healthy controls. At last, we made multivariate pattern analysis to differentiate rolandic epilepsy patients without IED from healthy controls.Results:We found significant differences among the three groups by one-way ANOVA in the right rectus gyrus (REC), right rolandic operculum (ROL), right thalamus (THA), left middle frontal gyrus (MFG), right postcentral gyrus (PoCG), and left superior parietal gyrus (SPG). Compared with healthy controls, we found distinguished intrinsic activity abnormality in thalamus, PoCG, ROL, REC, and SPG for IED group. In addition, we found distinguished intrinsic activity abnormality in MFG, REC, ROL and PoCG in non-IED group compared with healthy controls. Moreover, the ALFF of the right thalamus positively correlated with the number of IEDs observed during EEG-fMRI recordings. The ALFF value of the left SPG negatively correlated with performance IQ in the IED group (r=-0.4822, P=0.05) and all patients (r=-0.3299, P=0.0376), but not in the non-IED group. In addition, we found that patients with left IED and right IED had abnormal ALFF in right rolandic operculum and supramarginal gyrus compared with healthy controls. Using multivariate pattern classification analysis, we were able to differentiate rolandic epilepsy individuals without interictal epileptiform discharges (non-IED group) with healthy controls (92.16%accuracy, P<0.001).Conclusions:Rolandic epilepsy patients with or without interictal epileptiform discharges have intrinsic brain activity abnormalities in the right PoCG and ROL. The right thalamus may be associated with discharges of rolandic epilepsy. Intrinsic brain activity and multivariate pattern classification techniques are promising tools to differentiate rolandic epilepsy without interictal epileptiform discharges from healthy controls.更多还原