迷走神经刺激（Vagus nerve stimulation，VNS）是药物难治性癫痫（drug-resistant epilepsy，DRE）的一种有效疗法。但受限于深脑信号监测及分析的手段，VNS 疗法的抗癫痫机制尚未清晰，患者的筛选以及刺激参数的优化也缺乏客观的依据，导致部分患者在接受 VNS 治疗后无显著效果。针对这一问题，本文使用立体脑电图（stereo-electroencephalography，SEEG）技术，从致痫网络、重要核团以及脑网络拓扑三个层面探索了 VNS 对 DRE 患者脑网络急性影响的规律，主要开展了以下工作：首先，基于锁相值、相位滞后指数和加权相位滞后指数这三个同步性指标，研究了 VNS 疗法对 DRE 患者致痫网络同步性的急性影响的规律。通过在体实验发现，致痫网络的同步性指标随刺激电流的改变呈现不同的变化趋势。相较于其他两个同步性指标，加权相位滞后指数的同步性与临床疗效相关性最高。在临床最优参数下，疗效好的患者致痫网络呈现去同步化的趋势，且其同步性的基础值高于疗效差的患者，揭示了同步性指标用于预测 VNS 疗效的可能性。其次，使用功率谱密度的斜率探索了 VNS 疗法对 DRE 患者海马和杏仁核兴奋抑制性的急性调节影响。结果表明，海马和杏仁核的兴奋抑制失衡性随刺激电流的改变而改变，且两侧脑区对刺激参数的响应呈现不同的趋势。分析了在临床最优参数下 VNS 疗法对于不同疗效患者的影响，证实了 VNS 疗法对患者海马和杏仁核的兴奋抑制失衡性的影响与疗效之间呈正相关特性。最后，使用加权相位滞后指数构建了 15 名 DRE 患者的深部脑网络，探索了VNS 疗法对脑网络拓扑属性的急性影响。发现了不同受试者对 VNS 刺激的显著响应出现在不同的刺激强度范围内，这些显著响应的参数可能成为临床参数调节的重要参考。分析了在临床最优参数下，DRE 患者深部脑网络的特征路径长度，全局效率、聚类系数和局部效率的变化趋势，实验结果证实了 VNS 的抗癫痫效果与急性刺激下的脑网络拓扑属性影响相关。综上所述，本研究为确定 VNS 治疗癫痫的潜在标志物、患者筛选以及治疗参数选择提供了支撑证据。同时，本研究为采用 SEEG 监测及分析技术探索神经系统疾病的治疗机制提供了重要支持，具有重要的临床和科学意义。
Vagus nerve stimulation (VNS) is an effective therapy for drug-resistant epilepsy (DRE). However, limited by the means of deep brain signal monitoring and analysis, the antiepileptic mechanism of VNS therapy has not yet been clarified, and there is a lack of objective basis for patient screening and optimization of the stimulation parameters, resulting in no significant effect in some patients after receiving VNS therapy. To address this issue, we used stereo-electroencephalography (SEEG) to explore the acute effects of VNS on the brain network of DRE patients from three levels: epileptogenic network,important nuclei in the vagal upstream network and brain network topology. The main research works of this thesis are as follows:First, based on three synchronization indices, namely phase-locked value, phaselag index and weighted phase-lag index, the pattern of acute effects of VNS therapy on the synchronization of epileptogenic network in DRE patients was investigated. The invivo experiments revealed that the synchronization indexes of the epileptogenic network showed different trends with the change of stimulation intensity. Compared with the other two synchronization indices, the synchronization based on weighted phase lag index had the highest correlation with clinical efficacy. Under the clinically optimal parameters, the epileptogenic network of patients with good efficacy showed a trend of desynchronization, and the basal value of its synchronization was higher than that of patients with poor efficacy, revealing the possibility of the synchronization index to be used for predicting the efficacy of VNS.Second, the effects of VNS therapy on the acute modulation of excitatory-inhibitory properties of the hippocampus and amygdala, important nuclei in the vagal upstream network in patients with DRE, were explored using the slopes of the power spectral densities. The results showed that the excitation-inhibition imbalance in the hippocampus and amygdala changed with the stimulation intensity and that the two brain regions showed different trends in response to the stimulation parameters. The positive correlation between the effects of VNS therapy on the excitatory-inhibitory imbalance in the hippocampus and amygdala of patients and their efficacy was confirmed by the analysis of the effects of VNS therapy on patients with different efficacies under clinically optimal parameters.Finally, the acute effects of VNS therapy on brain network topological properties were explored by constructing deep brain networks of 15 DRE patients using weighted phase lag indices. It was found that significant responses to VNS stimulation appearedat different stimulion intensity ranges in different subjects, and the parameters of these significant responses may become important references for the regulation of clinical parameters. The trends of the characteristic path lengths, global efficiency, clustering coefficients and local efficiency of the deep brain network in DRE patients were analyzedunder the clinically optimal parameters, and the experimental results confirmed that the antiepileptic effect of VNS was related to the effect of brain network properties under acute stimulation.In summary, this study provides theoretical support for identifying potential biomarkers for VNS treatment of epilepsy, patient screening, and selection of therapeutic parameters. Meanwhile, this study provides important support for exploring the therapeutic mechanisms of neurological disorders using SEEG monitoring and analysis techniques,which is of great clinical and scientific significance.
