In an animal study, hyper-ammonemia results in increased numbers of swollen astrocytes, increased immuno-reactivity of glutamine synthetase, and some cytoskeletal proteins like the intermediate filament glial fibrillary acidic protein. These increases in swollen astrocytes and GFAP immuno-reactivity can be reduced by GS inhibition. The GFAP is expressed in astrocytes and is involved in cell structure maintenance and functioning of the blood brain barrier. It is also proposed to play a role in astrocyteneuron interactions, which may explain the alteration of astrocyte morphology and function. However, the limited correlation between decreased DMN integrity and ammonia level in this and another small sample-sized studies also imply a more complicated process in HE development or an adaptive change of functional LEE011 network that exists in each individual subject with a diverse clinical profile. Our results reveal significant antero-posterior functional disconnection in the PCC functional maps. Similar results have been suggested in subjects with increasing depth of sleep, wherein the posterior areas strengthen their connectivity, while connections between the frontal and posterior regions are lost. Functional uncoupling between the PCC and ACC and MFC may impair the brain’s ability to integrate information. Results of the present study suggest that integrated DMN activity may reflect ongoing conscious mental activity. In addition, increased MD, especially in the ACC, has good correlation with the disruption of the antero-posterior functional network, which further supports the notion that the PCC/precuneus are the central core within the DMN. Early hepatic encephalopathy is not identified by structural abnormalities. Instead, a comprehensive neurologic examination is usually required. Revealing the subtle functional alteration in “sub-clinical HE” is even more difficult. In an FDG-PET study, mHE is associated with decreased glucose uptake and blood flow in the ACC, medial frontal region, and precuneus. These indicate that brain regions involved in controlling the “attention system” responsible for monitoring is less active in mHE patients than in normal subjects. The NP tests, especially the block design and digit-symbol tests, are attention-demanding. Taken together, these findings support the hypothesis that a part of DMN connectivity fluctuates as it is more closely related to the presence of cognitive processes and can be altered earlier in liver cirrhosis. This study has some limitations. In subject recruitment, significantly ill patients were excluded from MRI study for safety concerns. As such, the preservation of DMN in comatose patients is unknown. Moreover, cirrhotic patients with altered consciousness are difficult to hold motionless during MRI scanning. Since the sample size is small, a relative low threshold for head motion criteria in data analysis has been applied, which might affect the results. Further studies with larger sample populations are warranted to assess the effect of therapy on functional network. Further validation of this diagnostic value on mHE by rs-fMRI is also required. In conclusion, HE patients show connectivity de-coupling between the fronto-posterior areas of DMN, which is associated with the degree of HE and brain oedema. Rs-fMRI can be used to investigate variability among patients with differing symptom profiles, including sub-clinical states.