Gray matter abnormalities in pediatric focal and generalized epilepsy: A voxel-based morphometry study with MRI at 3.0 T

Background : Our aims were to explore gray matter volume (GMV) abnormalities in focal and generalized epilepsy children compared to healthy group using a voxel-based morphometry (VBM) methodology, and to determine whether the regions of any observed structural changes are correlated with clinical characteristics. Methods: Thirty-two epilepsy children ( 18 with focal epilepsy, 14 with generalized epilepsy) and 18 control subjects were recruited. All participants were imaged structural MRI scans with a 3.0 T MR system. VBM analysis using statistical parametric mapping 8 (SPM8) was first performed to detect gray matter reduction in focal and generalized epilepsy children compared to controls. Then we analyzed the correlations between GMV changes and clinical features of epilepsy patients. Results: In the patient-control group comparison of VBM, the left hippocampus, left thalamus, left paraHippocampal and bilateral putamen exhibited a significantly GMV decreased. In stratified analysis, the focal epilepsy children showed GMV reductions in left thalamus, left hippocampus and bilateral putamen relative to controls; the GMV exhibited extensive reductions in left parahippocampal, left putamen, and right supplementary motor area in generalized epilepsy children compared to controls. No significant difference was found in regions of GMV abnormalities correlated with disease durationand age of onset in both the focal and generalized epilepsy groups. Conclusions: By performing VBM to detect GMV changes in children, our study demonstrates the GMV reductions exist in different regions in focal and generalized epilepsy patients. Our study further provides structural neuroimaging evidence on the pathophysiology of focal and generalized epilepsy children.

. Epilepsy can severely affect the development and function of immature brain, which may lead to cognitive or intellectual dysfunctions. The voxel-based morphometry (VBM), an automated technique for MRI analyses, is being increasingly applied in cerebral anatomical changes in epilepsy patients [3]. Through VBM emerged from recent developments in neuroimaging analysis techniques, investigators could explore gray matter volumes (GMV) abnormalities in epileptic patients to reveal the cause of the cognitive or intellectual dysfunctions.
Different pathophysiological mechanisms between focal epilepsy and generalized epilepsy have been explored in previous studies [5][6][7][8]. The study about interictal excitability differences demonstrate the impact of cortical excitatory/inhibitory function manifest bilateral in idiopathic generalized epilepsy, whereas it diffuse over the epileptic focus but remained lateralized in focal epilepsy [8]. Although some endeavor have been made in this filed, the pathophysiological basis differences are not fully understood yet. In addition, there are less study using VBM to explore GMV abnormalities in focal and generalized epilepsy compared to control, especially within the scope of children.
Epilepsy is the most common neurological condition in children and the characteristics of high prevalence, high morbidity, and high costs can result in heavy burden on family and society [9]. The prevalence of epilepsy in children ranges from 3.2-5.5/1,000 in developed countries to 3.6-44/1,000 in underdeveloped countries [10]. The social problems deriving from children epilepsy may be worse in our country [11]. The prevalence of children lifetime epilepsy is 4.5‰ aged 0 to 4 years old, 5.34‰ aged 5 to 9 years old, 6.20‰ aged 10 to 14 years old in china [12]. The prevalence of children lifetime epilepsy gradually increases with age. There are a small number of studies concentrating on the patients under 3 years old , because previous studies demonstrated a significantly improved seizure outcome compared to rates in older cohorts [13][14][15]. Though the prevalence of children lifetime epilepsy in older children forms a large proportion, they get less focus.
And it is more meaningful using older children as study population when exploring the relationship between age at seizure onset or epilepsy duration and changes in GMV.
Here, we investigated the regions of GMV abnormalities in focal and generalized epilepsy compared to the controls respectively. In addition, we correlated the volume changes of several brain regions with age at seizure onset and epilepsy duration. The aim of this study was to determine changes in GMV and associations with the development and progression of childhood epilepsy.

Subjects and clinical information
The study populations consisted of 32 right-handed epilepsy children aged 5 to 14 years old (mean age 8.6± 2.7 years, 24 males and 8 females). Epilepsy children were further divided into 18 right-handed children patients with focal epilepsy (mean age 8.6± 2.4 years, 14 males and 4 females) and 14 right-handed children patients with generalized epilepsy(mean age 8.6± 3.0 years, 10 males and 4 females). The diagnoses of epilepsy were determined in Guangzhou Women and Children's Medical center according to the criteria by the International League Against Epilepsy in 2017 [4]. All patients had had MRI scans and had long-term EEG records. The inclusion criteria were (1) diagnosed by an Epileptologist based on clinical, EEG and MRI findings; (2)age between 5 and 14 years; (3)attendance at standard schools; (4) high-resolution MR images acquired on the same type 3T research scanner. The exclusion criteria were (1) existence of organic brain disorder; (2) abnormalities in brain MRI scans; (3) 'Combined Generalized and Focal Epilepsy' and 'Unknown' were not included.
For group comparison, 18 healthy children (mean age 8.3± 1.8 years, 11 males and 7 females) who were similar in age, gender and education level were recruited as control group. Children with IQ < 70, a past history of neurologic or psychiatric disorders, a family history of epilepsy or a history of usage of medications damaging nervous system were excluded. This study was approved by the local ethics committee and informed consent was obtained for all participants from their parents.

MRI Acquisition
All of the participants underwent high-resolution brain MRI on a Siemens Trio 3T scanner.
For VBM analysis, high-resolution three-dimensional magnetization prepared rapid gradient echo (MPRAGE) sequence was acquired for each subject using the following parameters: TR = 1160ms, TE = 4.19ms, 3TI = 600ms, field of view = 140 × 250mm 2 , matrix size =256 × 192, slice thickness = 1.2mm, and flip angle = 15∘). During the process of scanning, all subjects were asked to be relaxed and keep their eyes closed but stay awake.

Voxel-based morphometry
The quantitative analysis of MRI was performed with the VBM8 Toolbox within SPM8 (Wellcome Trust Centre for Neuroimaging, https://www.fil.ion.ucl.ac.uk/spm/) running under MATLAB R2013b (The MathWorks, Natick, MA, USA). According the standard protocol, the VBM analysis comprised the following steps.
The quality of each image was inspected and the image origin was set to the anterior commissure. A customized template appropriate to the study population was created because of the wide range of age of participants. All native images were normalized to the same stereotactic space by registering each to the customized template image for spatial normalization. The normalized images were then segmented into gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF). Finally, the normalized, segmented GM images were smoothed with an isotropic Gaussian kernel with a sigma of 8 mm for statistical analyses. Finally, differences in cerebral GM density between the patient and control groups were evaluated using the voxelwise generalised linear model applied using permutationbased non-parametric testing (5000 permutations).

Statistical analyses
Prior to statistical analyses, differences in demographic were identified using an independent two-sample t-test for continuous variables and x 2 tests for dichotomous variables. The analyses were performed using SPSS version 22.0 (SPSS Inc., Chicago, IL, USA) and p values of <0.05 was considered indicative of statistical significance.
Using the general linear model, voxel-wise comparisons of GMV were performed between patients and controls, with age and gender as confounding covariates. The significance of group differences was estimated with the theory of p<0.001 uncorrected and the cluster size was set at >100 voxels.
To investigate the relationship between duration of epilepsy or age at epilepsy onset and GMV, the general linear model was applied. A statistical significance was accepted if p<0.05.

Demographic and clinical characteristics
The details of demographic and clinical characteristics for patients and healthy groups are showed in Table 1. Age and gender did not differ between the control and local epilepsy patients, generalized epilepsy patients or total epilepsy patients group (p>0.05). Also the specific diagnosis of epileptic syndromes among focal epilepsy and generalized epilepsy groups are listed in Table 1.

VBM analysis between control group and total epilepsy patients
Three clusters exhibited significant decreases in GMV in the whole brain VBM comparison between patients with epilepsy and control group. Within these clusters, the left hippocampus, left thalamus, left paraHippocampal and bilateral putamen were involved(Table 2, Figure 1).

VBM analysis between control group and focal epilepsy, generalized epilepsy patients
Compared to controls, patients with local epilepsy had significant GMV reductions in left thalamus, left hippocampus and bilateral putamen (p<0.001, uncorrected, Table 3, Figure   2). In patients with generalized epilepsy, the GMV was significantly decreased in left parahippocampal, left putamen, and right supplementary motor area (p<0.001, uncorrected, Table 4, Figure 3).

Correlations with seizure onset age and epilepsy duration
In whole-brain voxel correlation analysis, no correlations were found between GMV and the duration of epilepsy, age at epilepsy onset in the focal epilepsy and generalized epilepsy group (Figure 4, Figure 5).

Discussion
The pathophysiologic and anatomical mechanism of two different types of epilepsy, focal epilepsy and generalized epilepsy, remains inconclusive. The application of VBM provides an relatively unbiased tool to study the structural abnormalities of whole brain grey matter [16]. Although many previous studies have investigate abnormalities in grey matter in patients with the specific epilepsy syndromes [17][18][19][20][21], few studies have examine GMV difference according to the classification of epilepsy type, focal and generalized epilepsy.
To explore the serious effects of epilepsy on brain in incipient stage of life span, we used children as study objects.
This study attempted to explore the structural alterations of GMV in children with focal and generalized epilepsy, and to investigate the relationships between abnormal regions

Availability of data and materials
The datasets generated and/or analyzed during the current study available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
This work was supported by grants from Guangdong Key Project in "Development of new tools for diagnosis and treatment of Autism " (2018B030335001).     Grey matter abnormalities in focal epilepsy: cluster level significant regions(peak level p < 0.001 and cluster extent threshold =100 voxels). Abnormalities appear to be relatively decreased in left thalamus, left hippocampus and bilateral putamen.

Figure 3
Grey matter abnormalities in generalizes epilepsy: cluster level significant regions (peak level p < 0.001 and cluster extent threshold =100 voxels).
Abnormalities appear to be relatively decreased in left parahippocampal, left putamen, and right supplementary motor area.

Figure 4
Relationship between GMV and the duration of epilepsy, age at epilepsy onset in the focal epilepsy. Relationship between GMV and the duration of epilepsy, age at epilepsy onset in the generalized epilepsy.