Division of Brain, Imaging and Behaviour Systems, Krembil Brain Institute, Krembil Research Institute, Toronto, ON, Canada.
Institute of Medical Science, University of Toronto, Toronto, ON, Canada.
Pain. 2019 Jul;160(7):1670-1679. doi: 10.1097/j.pain.0000000000001545.
Therapeutic interventions for neuropathic pain, such as the N-methyl-D-aspartate (NMDA) antagonist ketamine, can vary widely in effectiveness. In this study, we conducted a longitudinal functional MRI study to test the hypothesis that the pain-relieving effect of ketamine is the result of reversal of abnormalities in regional low-frequency brain oscillations (LFOs) and abnormal cross-network functional connectivity (FC) of the dynamic pain connectome. We found that (1) ketamine decreased regional LFOs in the posterior cingulate cortex of the default mode network, (2) a machine-learning algorithm demonstrated that treatment-induced brain changes could be used to make generalizable inferences about pain relief, (3) treatment responders exhibited a significant decrease in cross-network static FC between the posterior cingulate cortex and regions of the sensorimotor and salience networks following treatment, (4) the degree of reduced cross-network FC correlated with the amount of pain relief, and (5) ketamine treatment did not produce significant differences in static or dynamic FC within the ascending nociceptive or descending antinociceptive pathway. These findings support the proposition that regional LFOs contribute to cross-network connectivity that underlie the effectiveness of ketamine to produce significant relief from neuropathic pain. Together with our recent findings that pretreatment dynamic FC of the descending antinociceptive pathway can predict ketamine treatment outcomes, these new findings indicate that pain relief from ketamine arises from a combination of flexible pretreatment FC of the descending antinocieptive pathway together with plasticity (reduction) of cross-network connectivity of the default mode network with sensorimotor and salience networks.
治疗神经病理性疼痛的干预措施，如 N-甲基-D-天冬氨酸（NMDA）拮抗剂氯胺酮，其疗效差异很大。在这项研究中，我们进行了一项纵向功能磁共振成像研究，以检验以下假设：氯胺酮的镇痛作用是由于逆转局部低频脑振荡（LFO）和动态痛连接组的异常跨网络功能连接（FC）的异常。我们发现：（1）氯胺酮降低了默认模式网络后扣带回皮质的局部 LFO；（2）机器学习算法表明，治疗引起的大脑变化可用于对疼痛缓解进行可推广的推断；（3）治疗反应者在治疗后表现出后扣带回皮质与感觉运动和突显网络区域之间的跨网络静态 FC 显著降低；（4）跨网络 FC 的减少程度与疼痛缓解的程度相关；（5）氯胺酮治疗不会导致上行伤害感受或下行抗伤害感受通路中的静态或动态 FC 产生显著差异。这些发现支持了这样一种观点，即局部 LFOs 有助于跨网络连接，这是氯胺酮产生显著缓解神经病理性疼痛的有效性的基础。结合我们最近发现的下行抗伤害性通路的预处理动态 FC 可以预测氯胺酮治疗结果，这些新发现表明，氯胺酮的镇痛作用源于下行抗伤害性通路的预处理 FC 的灵活性与默认模式网络与感觉运动和突显网络之间的跨网络连接的可塑性（减少）的组合。
