The goal of this study was to characterize cerebral cortex thickness patterns in juvenile myoclonic epilepsy (JME). Surface-based morphometry (SBM) was applied to process brain magnetic resonance images acquired from 24 patients with JME and 40 healthy controls and quantify cerebral cortex thickness. Differences in cortical thickness between patients and controls were determined using generalized linear model (covariates: age and gender). In patients with JME, thickness increase was detected bilaterally within localized regions in the orbitofrontal and mesial frontal cortices. Such thickness patterns coexisted with significant bilateral reduction in thalamic volume. These findings confirm that the underlying mechanisms in JME are related to aberrant corticothalamic structure and indicate that frontal cortex abnormalities are possibly linked to regional increase in cerebral cortical thickness.
Juvenile myoclonic epilepsy (JME) is a well-defined idiopathic generalized epilepsy (IGE) syndrome that is characterized by age-related onset of seizures manifesting as myoclonic jerks, generalized tonic–clonic seizures, and sometimes typical absence seizures (Pedersen & Petersen, 1998). Clinically, patients reveal no focal neurologic abnormalities and standard electroencephalography (EEG) recordings often show bilaterally diffuse generalized spike-wave (GSW) or polyspike-wave activity, sometimes with frontocentral predominance (Pedersen & Petersen, 1998). Quantitative neuroimaging studies of patients have detected several subtle neuroanatomic abnormalities, including thalamic gray matter volume (GMV) reduction and increased ventromedial frontal cortex (Duncan, 2005). These observations suggest that the underlying mechanisms in JME are related to aberrant frontothalamic structure (Duncan, 2005).
Recently, we investigated cerebral cortex surface morphology in a group of JME patients and reported anomalies in a number of regions dispersed throughout the cerebral cortex (Ronan et al., 2012). These morphologic alterations spared the frontal cortex and were related predominantly to changes in cortical surface area. Such cortical surface area–based findings, which likely reflect relevant JME-related pathologic processes, did not relate to previously described frontal cortex GMV alterations (Woermann et al., 1999).
Building on our previous investigation of cortical surface morphology, in this study, we examined cerebral cortex thickness patterns in the same patient cohort.
The medical research ethics committee at Beaumont Hospital Dublin approved this study. Written informed consent was obtained from all participants.
In total, 24 patients with JME were recruited from the epilepsy clinic at Beaumont Hospital and considered for this study. This sample of patients was described in detail in our previous report (Ronan et al., 2012). All patients underwent comprehensive evaluation that confirmed the clinical features of JME as defined by the International League Against Epilepsy (ILAE). In addition, 40 control subjects with no known neurologic deficits were included. See Table 1 for a summarized description of study participants.
Table 1. Demographics and MRI-derived brain structural parameters for study participants
SD, standard deviation.
Age at seizure onset and epilepsy duration are reported in years. Volume values are reported in mm3, surface area in mm2, and thickness in mm.
Intracranial volume (ICV) was included as covariate in the comparisons between patients and controls.
Patients and controls underwent brain magnetic resonance imaging (MRI) using a 1.5 T scanner (Signa, GE, Milwaukee, WI, U.S.A.) at Beaumont Hospital. A three-dimensional T1-weighted spoiled gradient recalled sequence [TR/TE = 10.1 msec/4.2 msec, TI = 450 msec, flip angle = 20 degrees, field of view (FOV) = 24 ×24 cm, matrix = 256 × 256] with 124 sagittal slices (slice thickness = 1.5 mm) was used to acquire the images.
MR images were processed using FreeSurfer (version 4.50, https://surfer.nmr.mgh.harvard.edu), a fully automated surface-based morphometry (SBM) image analysis tool. A detailed description of FreeSurfer process can be found elsewhere (Dale et al., 1999). Here, we applied FreeSurfer to reconstruct cortical surfaces as described previously (Ronan et al., 2012).
MRI data analysis and statistics
At each vertex, cerebral cortex thickness was quantified as the average of the shortest distance between the gray–white matter surface and the gray matter–cerebrospinal fluid (CSF) surface (Fischl & Dale, 2000).
Differences in cortical thickness between patients and controls were analyzed by computing a generalized linear model (GLM) of the effect of case–control status at each vertex across the entire cortex (covariates: age and gender).
Correlation of cortical thickness with thalamic volume and clinical variables
To examine the relationship between hemispheric cortical thickness and thalamic structure, the volume of the thalamus was calculated using FreeSurfer. A linear regression model (covariates: age and gender) was employed to assess the correlation between thalamic volume and cortical thickness. In addition, separate linear regression models were applied to assess the correlation of cortical thickness with age at seizure onset and duration of epilepsy.
The data was smoothed with a 15-mm full-width half maximum (FWHM) Gaussian kernel. Statistical parametric maps of significant group difference and correlations were corrected for multiple comparisons to control for false-positive results using Monte Carlo permutation cluster analysis (10,000 permutations) at a threshold of p<0.05. This method is less conservative compared to other correction methods.
Previously we reported no significant differences between the patient and control groups in age, gender, or intracranial volume (ICV). Furthermore, no significant differences were noted in total cerebral cortex volume, surface area, or average thickness (Ronan et al., 2012). See Table 1 for further details.
Group difference in cerebral cortex thickness
Figure 1 illustrates the results of the cerebral cortex thickness analysis. Increased cortical thickness was detected bilaterally in patients with JME relative to controls within localized regions in the orbitofrontal cortex and the medial side of the superior frontal gyrus. Further, increased cortical thickness was detected within regions in the right precuneus and inferior parietal cortices, and left inferior and superior temporal gyri.
Correlation of cortical thickness with thalamic volume and clinical variables
Compared to controls, patients with JME displayed significant bilateral reduction in thalamic volume (see Table 1). In each hemisphere, thalamic volume was found to correlate negatively with the thickness of the orbitofrontal cortex. Negative correlation was also noted between orbitofrontal cortical thickness and age at seizure onset. Duration of epilepsy correlated negatively with the thickness of the insular cortex.
To our knowledge, this study is the first to present evidence of regional increase of cerebral cortex thickness in JME. Localized increase in cortical thickness was evident in patients relative to controls within the ventromedial frontal cortex bilaterally as well as a number of regions within the temporal and parietal cortices. These findings highlight significant neuroanatomic traits that may represent “intermediate” phenotypes relating to the underlying causes of this epilepsy syndrome.
The pathophysiologic mechanisms in JME have traditionally been linked to frontally predominant abnormal corticothalamic networks. Neuroimaging studies of patients have reported bilateral frontothalamic neuroanatomic and functional alterations indicating aberrant frontothalamic structure (Duncan, 2005). Using voxel-based morphometry (VBM), several investigators reported increased GMV in ventromedial frontal cortex and bilateral reduction in thalamic volume (Woermann et al., 1999; Betting et al., 2006). These neuroanatomic quantitative traits appeared specific to patients with JME (Betting et al., 2006); however, their significance and relationship to the underlying mechanisms remain unclear.
It has recently been illustrated that the examination of cortical surface area and thickness is superior to GMV when assessing cerebral cortex features (Panizzon et al., 2009). Each of these cerebral cortex traits contributes independently to GMV and reflects on distinct neurobiologic and genetic mechanisms (Panizzon et al., 2009). Studies of cerebral cortex alterations in other brain-related conditions revealed that abnormalities in GMV may relate to alterations in either thickness or surface area with minimal spatial overlap between thickness and surface area abnormal patterns (Ecker et al., 2013). Based on the same JME patient sample studied in this investigation, we recently reported widespread cortical surface area changes involving several cortical regions outside the frontal lobe (Ronan et al., 2012). These surface area–based cerebral cortex changes, which may indicate altered surface folding, were distinct from the thickness patterns detected in the present study. The identified alterations in orbitofrontal and mesial frontal cortical thickness support previously described frontal GMV abnormalities (Woermann et al., 1999; Betting et al., 2006), and indicate that such GMV changes are likely explained by an increase in cortical thickness.
Holmes et al. (2010) applied dense array scalp electroencephalography (EEG) recordings to study the epileptiform discharges in 10 patients with JME and found all patients to have epileptiform activity originating from the orbitofrontal and medial frontopolar cortices (Holmes et al., 2010). The frontal cortex thickness patterns identified in the present study may represent neuroanatomic correlates of these electrophysiologic observations. This speculation is supported by the significant negative correlation we noted between orbitofrontal cortical thickness and age at seizure onset, which emphasizes the relevance of orbitofrontal cortex structural alterations to JME underlying mechanisms. The observed correlation between orbitofrontal cortical thickness and thalamic volume also highlights the relevance of abnormal frontothalamic interactions in this epilepsy syndrome. Previously identified mutations in the EFHC1 gene are believed to interfere with embryonic cortical development and contribute to cortical dysplasia (de Nijs et al., 2009), suggesting that malformations of cortical development (MCDs) caused by certain genetic factors can, at least partially, explain the underlying mechanisms in JME (Wong, 2010). The identified localized regions of increased cortical thickness in this study may therefore reflect underlying processes related to genetic or early developmental factors.
In summary, this study revealed evidence of increased thickness in regions within the ventromedial frontal cortex associating with bilateral thalamic volume loss in patients with JME. These findings support the view that this epilepsy syndrome is related to aberrant corticothalamic structure.
The authors thank all patients and participants who took part in this study. This work was funded by a Health Research Board fellowship (HSR/2006/7).
None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.