In recent stroke studies, connectivity-based approaches have been used to investigate recovery-related indicators. However, most of previous studies were performed in heterogenous patients with different lesion locations or different types of stroke. In addition, there is not enough investigation on recovery-related network analysis in the infratentorial stroke (ITS), which may show different network dynamics from the supratentorial stroke (STS). In this study, we investigated the differences in alteration of motor network between STS and ITS in ischemic stroke patients.

Forty first-onset ischemic stroke patients were recruited within 2 weeks after stroke onset. Patients were divided into STS and ITS groups according to their lesion location. All patients underwent resting-state fMRI scans twice (2 weeks and 3 months after onset). Twenty-four healthy subjects participated as an age-matched control group. To investigate the altered connectivity during recovery and to compare between groups, various analysis methods such as interhemispheric connectivity, network symmetry, and graph theoretical analysis which were already established in the previous studies were used.

In STS group, interhemispheric connectivity was significantly decreased and network symmetry was disrupted compared to control group at 2 weeks post-stroke, while the ITS group did not show differences compared to control group at this time point. During recovery, global efficiency and global reorganization including increasing network efficiency and randomization showed only in the STS group. In contrast, the ITS group demonstrated significantly increased interhemispheric connectivity during recovery. There was an interactive relationship between cortico-cerebellar connectivity and interhemispheric connectivity in the ITS group.

Alterations of recovery-related motor network connectivity in ITS patients were very different from that of STS patients. These results may be caused by differences of damage of diverse motor-related pathways by stroke lesion.