Multivariate Granger Causal Analysis Of The Response Of Resting Cerebral Network To Stroke Hemiplegia By Acupuncture At Yanglingquan

ObjectiveIn the current study, taking Yanglingquan (GB34) as a cut-in point, we tried to detect the responsive characteristics of effective connectivity between the resting-state networks in stroke patients with hemiplegia by acupuncture at Yanglingquan (GB34) from the aspects of multiple resting-state brain networks, in order to provide data support for promoting brain function plasticity by traditional Chinese medicine methods.MethodsA total of 19 stroke patients with hemiplegia and another 17 age- and gender-matched healthy control subjects were recruited in the current study. Stroke patients were scanned in resting state and acupuncture procedure respectively, while the healthy controls only scanned in resting state. We extracted multiple resting-state networks by the method of independent component analysis (ICA), and then explored the characteristics of abnormal changes of brain functional network in stroke patients with hemiplegia by contrastive analysis of the changes of functional connectivity within resting-state networks between stroke patients and healthy subjects. On this basis, the modulatory effects of acupuncture between multiple brain networks in stroke patients was further discussed by the method of Granger causality analysis (GCA).Results1.Based on ICA method, a total of 7 resting-state networks were extracted including left frontoparietal network (LFPN), right frontoparietal network (RFPN), anterior default mode network (aDMN), posterior default mode network (pDMN), sensorimotor network (SMN), visual network (VN), salience network (SN).2.Altered functional connectivity within resting-state networks in stroke patients with hemiplegia:Compared with healthy subjects, the functional connectivity within resting-state networks showed significantly enhanced or reduced. Among of them, these brain regions presenting strengthen mainly located in the lesion side, only a small amount of the brain regions of LFPN and aDMN in contralateral brain. However most other regions of brain network exhibited attenuated. The same brain region in different networks showed different changes in brain functional connectivity.3.Altered effective connectivity between resting-state networks in stroke patients with hemiplegia:Compared with healthy subjects, stroke patients with hemiplegia presented the more complex effective connectivity between brain networks, there were significant difference in causal links between SN and SMN, and between RFPN and aDMN (P<0.05).4.The regulating effect of acupuncture on resting-state brain networks in stroke patients: Among these causal connection between brain networks of apoplexy hemiplegia, LFPN inputted most information from other networks while DMN outputted most information to other networks before acupuncture, however the above results were reversal by acupuncture, LFPN outputted most information while DMN inputted most information, details were as follows:acupuncture reversed the direction of effective connectivity between LFPN and aDMN, pDMN, SN, and turned over the directon of effective connectivity between pDMN and LFPN, RFPN, SN. Meanwhile, these effective connectivities between LFPN and RFPN, SMN, between aDMN and pDMN, VN decreased while those between LFPN and VN, aDMN, between aDMN and SMN, between SMN and SN, between RFPN and VN increased. After acupuncture stimulation, LFPN outputted information to aDMN, then aDMN outputted the information to SMN, which formed the part of the information transmission loop. These causal relationship existed significant differences between LFPN and aDMN, VN and SMN, pDMN and LFPN, SN and LFPN (P<0.05) before and after acupuncture intervention.ConclusionStroke patients with hemiplegia showed the changes including enhancement and diminishment in functional connectivity of multiple resting-state networks, to a certain extent, which reflected the coexistence of impairments and compensation in brain networks. The characteristics of abnormal variations were present not only intranetwork but also internetwork. That the same brain region in different networks showed different changes in brain functional connectivity, may be the basis of effective connectivity between resting state networks. Acupuncture can regulate multiple brain networks, and probably integrate the effective connectivity internetwork by transferring information between advanced cognitive network and SMN by DMN as the relay station.