[Anterior and posterior default mode networks impairments in minimal hepatic encephalopathy]

To investigate the functional connectivity (FC) impairments within anterior and posterior default mode networks (aDMN and pDMN) for cirrhosis with minimal hepatic encephalopathy (MHE), and to analyze relationship between DMNs alterations with neurocognitive impairments.

From March 2009 to December 2011, a total of 43 cirrhotic patients without MHE (NMHE groups), 32 cirrhotic patients with current MHE (MHE groups), and 21 healthy controls were recruited in this study. All resting-state fMRI datasets were preprocessed and normalized into standard brain space. Two universal templates for aDMN and pDMN were applied to generate spatial patterns for aDMN and pDMN for each participate by dual regression. One sample t test were analyzed within groups, and one way ANOVA (analysis of variance) were performed to calculate the different FCs within aDMN and pDMN among groups. Additional correlation analysis was performed between different FCs within two DMNs and behavior scores.

Within aDMN, FC of MHE patients were significantly deceased than controls and NMHE groups in right superior frontal gyrus (SFG)/ bilateral medial frontal gyrus (MFG); while within pDMN, FC of MHE patients were significantly increased than controls and NMHE groups in right superior temporal gyrus (STG)/ middle temporal gyrus (MTG) and left cuneus/ bilateral calcarine gyri (one way ANOVA, P<0.01, cluster sizes>40, AlphaSim correction). The FC alterations in those region were significantly correlated with neurocognitive test scores (Pearson correlation coefficient, P<0.05).

Dual regression based aDMN and pDMN method was firstly applied to investigated functional connectivity impairments in MHE patients' brain. Functional deficits were endogenous within aDMN, while functional connectivity was compensatory enhancements within pDMN. These compensatory enhancements indicated that patients with current MHE had the potential to additionally recruit more neurological resource to accomplish the specific actions. Therefore, the alterations of dual regression based aDMN and pDMN can be potential neuroimaging biomarkers for MHE studies.