Abstract
Hong S-J, Bernhardt BC, Gill RS, Bernasconi N, Bernasconi A. Brain 2017;140:2133–2143, https://doi.org/10.1093/brain/awx145.
Neuroimaging studies of malformations of cortical development have mainly focused on the characterization of the primary lesional substrate, while whole-brain investigations remain scarce. Our purpose was to assess large-scale brain organization in prevalent cortical malformations. Based on experimental evidence suggesting that distributed effects of focal insults are modulated by stages of brain development, we postulated differential patterns of network anomalies across subtypes of malformations. We studied a cohort of patients with FCD-II (n = 63), subcortical nodular heterotopia (n = 44), and polymicrogyria (n = 34), and compared them to 82 age- and sex-matched controls. Graph theoretical analysis of structural covariance networks indicated a consistent rearrangement towards a regularized architecture characterized by increased path length and clustering, as well as disrupted rich-club topology, overall suggestive of inefficient global and excessive local connectivity. Notably, there was a gradual shift in network reconfigurations across subgroups, with only subtle changes in FCD-II, moderate effects in heterotopia and maximal effects in polymicrogyria. Analysis of resting state functional connectivity also revealed gradual network changes, with most marked rearrangement in polymicrogyria; contrary to findings in the structural domain, however, functional architecture was characterized by decreases in both local and global parameters. Diverging results in the structural and functional domain were supported by formal structure-function coupling analysis. Our findings support the concept that time of insult during corticogenesis impacts the severity of topological network reconfiguration. Specifically, late-stage malformations, typified by polymicrogyria, may selectively disrupt the formation of large-scale cortico-cortical networks and thus lead to a more profound impact on whole-brain organization than early stage disturbances of predominantly radial migration patterns observed in cortical dysplasia type II, which likely affect a relatively confined cortical territory.
Commentary
Advances in MRI technology enable evaluation of global changes in both structure and function of the brain. The article by Hong et al. uses an interesting approach to the study of patients with malformations of cortical development (MCDs) and chronic epilepsy, using both structural and functional MRI.
Past quantitative evaluation of structural MRI shows significant changes in both focal and generalized epilepsies. Examples of structural brain changes in focal epilepsies include temporal lobe epilepsy caused by mesial temporal sclerosis, which shows volume and morphologic changes that are most pronounced in the hippocampus and mesial temporal structures but also involve the thalamus and neocortex in the cingulate and frontocentral regions (1). In focal frontal lobe epilepsy due to focal cortical dysplasia (FCD), subjects with FCD type I showed multilobar cortical thinning, which is most pronounced in ipsilateral frontal cortices, while subjects with FCD type II showed cortical thickening in temporal and postcentral cortices (2). In a recent review of structural MRI changes in genetic generalized epilepsy (GGE), there were significant structural differences between those with GGE and healthy controls with decreased volumes in the whole brain, thalamus, putamen, caudate, pallidum, and supplementary motor area (3).
Therefore, quantitative structural MRI studies stretch the limits of our understanding of focal and generalized epilepsy. For focal epilepsy, there tend to be accentuated structural changes in the region of seizure onset. However, there are also global, more generalized structural changes. On the other hand, while GGE seizures conceptually involve broad, bilateral hemispheric onset (4), quantitative structural MRI studies show relatively localized structural changes compared with controls. The underlying pathophysiology responsible for structural changes in epileptic seizures is multifactorial but includes the underlying etiology of the epileptic seizures, making the study of MCDs an important topic.
Functional MRI studies in persons with chronic epilepsy may also demonstrate focal and generalized changes. Using resting state blood-oxygenation level dependent functional magnetic resonance imaging (BOLD-MRI) in subjects with temporal lobe epilepsy compared with controls, previous studies have demonstrated increased intrahemispheric functional connectivity ipsilateral to the seizure focus but decreased interhemispheric functional connectivity. For example, several regions in the affected temporal lobe showed increased functional coupling with the ipsilateral insula and immediately neighboring subcortical regions. However, there was significant decreased functional connectivity between regions in the affected temporal lobe and contralateral homologous counterparts (5). In GGE, BOLD-MRI techniques showed altered connectivity in several cortical and subcortical regions, including the mesial frontal cortex, putamen, thalamus, and amygdala, supporting the hypothesis that specific regions play an important functional role in GGE (6).
Given this background, the article by Hong et al offers an insightful approach to the relationship of structural and functional MRI changes in subjects with MCD. Using associations of stages of neocortical development with different MCDs, the authors compare subjects with cortical dysplasias, heterotopias, and polymicrogyria using structural covariance assessments combined with resting-state functional MRI connectivity analysis. The authors studied 154 patients with MCD to demonstrate large-scale structural and functional network changes in subjects who show disruptions at different main stages of cortical development.
A key concept of their study is the correlation of dysfunction of specific neurodevelopmental stages with different MCDs. The first group of malformations includes those caused by abnormal neuroglial proliferation, which is represented in this study by FCD type II (FCD-II) and is typically characterized by cortical thickening on structural MRI. The second group is caused by disruption of neuronal migration, represented by subcortical nodular heterotopias in the study. Structural MRI findings in this group are grey matter nodules within various depths of the periventricular white matter. The third group is associated with postmigratory disturbances of cortical development, resulting in an excessive number of shallow gyri and sulci. This group is represented by polymicrogyria in the study. The authors hypothesized that the severity of structural and functional connectivity abnormalities would correlate with the timing of neurodevelopmental disruption, with malformations occurring in earlier neurodevelopmental stages (represented by FCD-II) resulting in mild abnormalities, those occurring in later neuromigrational stages (represented by subcortical nodular heterotopias) resulting in moderate changes, and those in postmigrational stages (represented by polymicrogyria) resulting in the most marked changes.
To define regions for analysis, the authors used automatic anatomical labeling to parcellate the neocortex into 78 regions (39 for each hemisphere). The investigators used cortical thickness measurements to define structural changes and BOLD-MRI to define functional changes. They used standardized techniques to compute clustering coefficient and characteristic path length in controls and MCD groups and used the rich-club organization phenomenon to document the connectedness between nodes.
For structural analysis, the networks in controls showed a pattern of interregional correlations characterized by strong connections within one hemisphere and moderate connectivity between hemispheres. In FCD II, there was both increased and decreased covariance compared with findings in controls, without a specific hemispheric pattern. However, heterotopias showed increased intrahemispheric covariance and decreased interhemispheric covariance. Polymicrogyrias showed a more marked pattern of densely increased intrahemispheric and almost absent interhemispheric covariance. Findings of increased path length and clustering, as well as disrupted rich-club topology, were overall suggestive of inefficient global and excessive local connectivity.
For functional analysis, again a gradual progression of abnormalities was observed when patients with MCD were compared with controls, with greatest anomalies in polymicrogyrias. However, as opposed to structural covariance, inter-regional functional connectivity measures showed an overall decrease, both within and across hemispheres, indicating both local and global inefficiencies. Differences in the structural and functional changes were supported by formal structure–function coupling analysis.
The findings in this study are interesting as they show a greater degree of both structural and functional MRI changes in MCDs, which are more prominent in MCDs associated with later neurodevelopmental stages. These findings confirm the initial hypothesis of a greater degree of abnormality in later-stage neurodevelopmental MCDs. However, there was a diffuse marked structure–function decoupling, especially in heterotopias and polymicrogyrias. Past studies of subjects with GGE epilepsy also show a structure-function decoupling, 6 which suggests that functional and structural MRI studies give differing and possibly complementary information about the underlying etiology and pathophysiology of epileptic seizures.
In general, the degree of functional and structural MRI abnormalities, as well as differences in the structural and functional changes as defined by structure–function coupling analysis, correlate with cognitive dysfunction of the MCDs. Overall, FCD, subcortical nodular heterotopias, and polymicrogyria show a continuum of severity of cognitive dysfunction, with polymicrogyria exhibiting the most severe abnormalities. The findings also highlight the pathophysiology of decoupling of connectivity between structural and functional MRI studies, suggestive that direct comparison of structural and functional connectivity changes in subjects with chronic epilepsy is important. Further study evaluating this decoupling relationship, especially for practical applications such as localization of seizures for epilepsy surgery, will be an important avenue of inquiry in the future.