Altered neuromagnetic activity in default mode network in childhood absence epilepsy

Abstract

Purpose: The electrophysiological characterization of resting state oscillatory functional connectivity within the default mode network (DMN) during interictal periods in childhood absence epilepsy (CAE) remains unclear. Using magnetoencephalographic (MEG) recordings, this study investigated how the connectivity within the DMN was altered in CAE.

Methods: Using a cross-sectional design, we analyzed MEG data from 33 children newly diagnosed with CAE and 26 controls matched for age and sex. The spectral power and functional connectivity of the DMN were estimated using minimum norm estimation combined with the Welch technique and corrected amplitude envelope correlation.

Results: Default mode network showed stronger activation in the delta band during the ictal period, however, the relative spectral power in other bands was significantly lower than that in the interictal period (p _corrected < 0.05 for DMN regions, except bilateral medial frontal cortex, left medial temporal lobe, left posterior cingulate cortex in the theta band, and the bilateral precuneus in the alpha band). It should be noted that the significant power peak in the alpha band was lost compared with the interictal data. Compared with controls, the interictal relative spectral power of DMN regions (except bilateral precuneus) in CAE patients was significantly increased in the delta band (p _corrected < 0.01), whereas the values of all DMN regions in the beta-gamma 2 band were significantly decreased (p _corrected < 0.01). In the higher frequency band (alpha-gamma1), especially in the beta and gamma1 band, the ictal node strength of DMN regions except the left precuneus was significantly higher than that in the interictal periods (p _corrected < 0.01), and the node strength of the right inferior parietal lobe increased most significantly in the beta band (Ictal: 3.8712 vs. Interictal: 0.7503, p _corrected < 0.01). Compared with the controls, the interictal node strength of DMN increased in all frequency bands, especially the right medial frontal cortex in the beta band (Controls: 0.1510 vs. Interictal: 3.527, p _corrected < 0.01). Comparing relative node strength between groups, the right precuneus in CAE children decreased significantly (β: Controls: 0.1009 vs. Interictal: 0.0475; γ 1: Controls:0.1149 vs. Interictal:0.0587, p _corrected < 0.01) such that it was no longer the central hub.

Conclusion: These findings indicated DMN abnormalities in CAE patients, even in interictal periods without interictal epileptic discharges. Abnormal functional connectivity in CAE may reflect abnormal anatomo-functional architectural integration in DMN, as a result of cognitive mental impairment and unconsciousness during absence seizure. Future studies are needed to examine if the altered functional connectivity can be used as a biomarker for treatment responses, cognitive dysfunction, and prognosis in CAE patients.

Keywords: childhood absence epilepsy; functional connectivity; ictal; interictal; magnetoencephalography; oscillations; spectral power.

FIGURE 1

Study profile.

FIGURE 2

Cortical activation. The distribution of the average cortical activation in control, interictal and ictal data. The activation maps are shown from the top and lateral views. Values range between 0 and 1, indicating the power of cortical signals relatively to the total signal power across the frequency spectrum. Only the cortical sources with a strength larger than 60% of the maximal value are displayed. The current strength for cortical sources is color-coded with large values represented in red.

FIGURE 3

The relative spectral power of each ROI within the DMN. The estimated normalized and averaged spectral power of 12 ROIs in six frequency bands across control, interictal and ictal data, respectively. δ, delta; θ, theta; α, alpha; β, beta; γ 1, gamma 1; γ 2, gamma 2. Left, left hemisphere. Right, right hemisphere.

FIGURE 4

The node strength of each ROI within the DMN. The node strength of each ROI from the delta to gamma 2 bands is color-coded for control, ictal and interictal data, respectively. The difference 1 (ictal minus interictal) and difference 2 (interictal minus control) of the node strength demonstrate the alterations in functional connectivity. A, IPL; B, MFC; C, MTL; D, PCu; E, PCC; F, LTL. *p < 0.05; **p < 0.01.

FIGURE 5

Results of MEG connectivity analysis. The histogram shows the relative node strength of 12 DMN regions in beta and gamma1 bands. The upper half represents the left hemisphere and the lower half represents the right hemisphere. All values are mean. ^aP < 0.01 compared to Ictal, ^bP < 0.01 compared to Interictal, and ^cP < 0.01 compared to controls.