Altered functional and effective connectivity in anticorrelated intrinsic networks in children with benign childhood epilepsy with centrotemporal spikes

Abstract

There are 2 intrinsic networks in the human brain: the task positive network (TPN) and task negative network (alternately termed the default mode network, DMN) in which inverse correlations have been observed during resting state and event-related functional magnetic resonance imaging (fMRI). The antagonism between the 2 networks might indicate a dynamic interaction in the brain that is associated with development.To evaluate the alterations in the relations of the 2 networks in children with benign childhood epilepsy with centrotemporal spikes (BECTS), resting state fMRI was performed in 17 patients with BECTS and 17 healthy controls. The functional and effective connectivities of 29 nodes in the TPN and DMN were analyzed. Positive functional connectivity (FC) within the networks and negative FC between the 2 networks were observed in both groups.The patients exhibited increased FC within both networks, particularly in the frontoparietal nodes such as the left superior frontal cortex, and enhanced antagonism between the 2 networks, suggesting abnormal functional integration of the nodes of the 2 networks in the patients. Granger causality analysis revealed a significant difference in the degree of outflow to inflow in the left superior frontal cortex and the left ventral occipital lobe.The alterations observed in the combined functional and effective connectivity analyses might indicate an association of an abnormal ability to integrate information between the DMN and TPN and the epileptic neuropathology of BECTS and provide preliminary evidence supporting the occurrence of abnormal development in children with BECTS.

Figure 1

The functional connectivity in patients BECTS (right) and healthy controls (left). The significant connections with P < 0.001 (FDR corrected for the multiple comparisons) are illustrated. The nodes in TPN are colored in red, and the nodes in DMN are colored in blue. The red lines delineate the positive functional connections (within network), and the blue lines represent the negative functional connections (between 2 networks). BECT = benign childhood epilepsy with centro-temporal spikes, FDR = False Discover Rate, DMN = Default Mode Network, TPN = Task Positive Network.

Figure 2

(A) The difference of functional connectivity between 2 groups. The red line represents the increased functional connections in patients compared with healthy controls, and the blue represents the decreased connections. Same as Figure 1, the red dots mean the nodes in TPN, and the blue one means the nodes in DMN. (B) The positive correlation between epilepsy duration (months) and the functional connection between SMA and right DLPFC. ^∗Residuals after controlling for the influence of the gender, age (years), and AEDs (linear regression with covariates including gender and age). The green arrowed line between Sections A and B means that the functional connection positively correlated with the duration of epilepsy.

Figure 3

The average Granger causality in patients with BECTS (right) and healthy controls (left). The arrows represent the direction of the effective connections. The nodes in TPN are colored in red, and the nodes in DMN are colored in blue. The arrowed line represents the direction of Granger causal connection between 2 nodes.

Figure 4

The within-group level causal flow (out–in degree). The red bars show the significant difference between groups (P < 0.05).