Frequency-Dependent Changes in Interhemispheric Functional Connectivity Measured by Resting-State fMRI in Children With Idiopathic Generalized Epilepsy

Abstract

Epilepsy is associated with abnormal spatiotemporal changes in resting-state brain connectivity, but how these changes are characterized in interhemispheric coupling remains unclear. This study aimed to characterize frequency-dependent alterations in voxel-wise mirrored homotopic connectivity (VMHC) measured by resting-state functional magnetic resonance imaging (rs-fMRI) in children with idiopathic generalized epilepsy (IGE). Rs-fMRI data were collected in 21 children with IGE and 22 demographically matched children with typical development. We used three resting-state frequency bands (full, 0.01-0.08 Hz; slow-4, 0.027-0.073 Hz; slow-5, 0.01-0.027 Hz) to compute VMHC and locate the significant foci. Voxel-wise p <0.001 and cluster-level p <0.05 cluster-level family-wise error correction was applied. In between-group comparisons, we identified that the full and higher frequency (slow-4) bands showed similar reductions in VMHC including Rolandic operculum, putamen, superior frontal, lateral parietal, middle cingulate, and precuneus in children with IGE. In the lower frequency band (slow-5), we identified specific reductions in VMHC in orbitofrontal and middle temporal gyri in children with IGE. Further analyses on main effects and interaction between group and frequency band suggested significant frequency-dependent changes in VMHC, and no significant interaction was found. The results were generally similar with global brain signal regression. Additional association analysis showed that VMHC in the putamen within the full and slow-4 bands was significantly positively correlated with chronological age in children with IGE, and the same analysis was non-significant in the controls; VMHC in the medial prefrontal region in the slow-4 band was significantly positively correlated with IQ performance sub-score. Our findings suggest that IGE children show frequency-dependent changes in interhemispheric integration that spans regions and systems involving cortical-subcortical, language, and visuomotor processing. Decreased functional coupling within the dorsal striatum may reflect atypical development in children with IGE.

Keywords: IGE; fMRI; frequency-dependent; functional connectivity; interhemispheric.

Figure 1

Between-group comparisons on VMHC. The three color labels represent statistically significant between-group differences, i.e., significantly decreased VMHC between the IGE and HC groups under the three frequency bands (red, typical band: 0.01–0.08 Hz; green, slow-4: 0.027–0.73 Hz; blue, slow-5: 0.01–0.027 Hz). The composed map reflects the topographical distribution of frequency-specific changes. These results were voxel-wise p <0.001, cluster level p <0.05, family-wise error (FWE) cluster-level corrected.

Figure 2

Main effects for group. Only significantly reduced VMHC was found (cold colorbar, IGE < HC). The results were voxel-wise p <0.001, cluster level p <0.05, family-wise error (FWE) cluster-level corrected. L, left; R, right.

Figure 3

Main effects for frequency band. Only significantly reduced VMHC was found (cold colorbar, slow-4 < slow-5). The results were voxel-wise p <0.001, cluster level p <0.05, family-wise error (FWE) cluster-level corrected. L, left; R, right.

Figure 4

Association analysis. Significantly positive associations were found between VMHC within the putamen under two frequency bands (full and slow-4) and chronological age and between VMHC within the medial prefrontal cortex and IQ performance subscore. (A) Putamen. (B) Medial prefrontal cortex.