Default mode network deactivation in pediatric temporal lobe epilepsy: Relationship to a working memory task and executive function tests

Abstract

Objectives: Children with temporal lobe epilepsy (TLE) exhibit executive dysfunction on traditional neuropsychological tests. There is limited evidence of different functional network alterations associated with this clinical executive dysfunction. This study investigates working memory deficits in children with TLE by assessing deactivation of the default mode network (DMN) on functional Magnetic Resonance Imaging (fMRI) and the relationship of DMN deactivation with fMRI behavioral findings and neuropsychological test performance.

Experimental design: fMRI was conducted on 15 children with TLE and 15 healthy controls (age: 8-16 years) while performing the N-back task in order to assess deactivation of the DMN. N-back accuracy, N-back reaction time, and neuropsychological tests of executive function (Delis-Kaplan Executive Function System [D-KEFS] Color-Word Interference and Card Sort tests) were also assessed.

Principal observations: During the N-back task, children with TLE exhibited significantly less deactivation of the DMN, primarily in the precuneus/posterior cingulate cortex compared with controls. These alterations significantly correlated with N-back behavioral findings and D-KEFS results.

Conclusions: Children with TLE exhibit executive dysfunction which correlates with DMN alterations. These findings suggest that the level of deactivation of specific functional networks may contribute to cognitive impairment in children with TLE. The findings also indicate that children with TLE have network alterations in extratemporal lobe brain regions.

Keywords: Default mode network; Executive dysfunction; N-back task; Neuropsychological testing; Pediatric; Temporal lobe epilepsy.

Fig. 1.

Patients with epilepsy show slower N-back reaction times compared with healthy controls with a trend towards lower N-back accuracy scores. *p < 0.05, §p < 0.1.

Fig. 2.

Children with epilepsy perform poorer on D-KEFS Color-Word Interference accuracy, D-KEFS Color-Word Interference speed performance, and D-KEFS Card Sort task accuracy compared with healthy controls, *p < 0.05.

Fig. 3.

Using bspmview software, this panel indicates the deactivation (negative contrast images, 0-back > 2-back). FWE cluster-level correction, cluster defining threshold p < 0.001, cluster size > 25 voxels. A - shows the controls only, the clusters of activation of the DMN are smaller and less intense compared to B during the N-back task. B - shows the patients with TLE only, the clusters of activation of the DMN are larger and more intense compared to A during the N-back task. C - shows the difference graph (TLE > controls contrast) indicating the regions that patients with TLE deactivate less during the N-back task. These significantly different regions are the left parietal lobe and the right precuneus, which are included in the DMN. The fMRI maps (in color) superimposed on anatomical images (gray scale). The red-yellow scale indicates intensity of activation of region. REF - results were confirmed using an a priori mask of the DMN map.