Diffusion abnormalities of the corpus callosum in patients with malformations of cortical development and epilepsy

Highlights

• We investigated diffusion properties of the CC in 32 MCD patients and 32 controls.

• Deterministic tractography and ROI-based analyses yielded FA, MD, λ|| and λ⊥.

• MCD patients presented significant reductions of FA, increases in MD and λ⊥.

• No differences between MCD subgroups, and no correlations with clinical parameters.

• MCD patients present widespread changes that extend beyond the macroscopic lesions.

Summary

Purpose

Diffusion tensor imaging (DTI) is a magnetic resonance imaging (MRI) technique that can characterize white matter (WM) architecture and microstructure. DTI has demonstrated extensive WM changes in patients with several epileptic syndromes, but few studies have focused on patients with malformations of cortical development (MCD). Our aim was to investigate the quantitative diffusion properties of the corpus callosum (CC), a major commissural bundle critical in inter-hemispheric connectivity, in a large group of patients with MCD.

Methods

Thirty-two MCD patients and 32 age and sex-matched control subjects were evaluated with DTI at 3.0 T. We analyzed the three major subdivisions of the CC (genu, body, and splenium) with deterministic tractography to yield fractional anisotropy (FA), mean diffusivity (MD), parallel diffusivity (λ||) and perpendicular diffusivity (λ⊥). We further assessed the CC with region of interest (ROI)-based analyses and evaluated different subgroups of MCD (polymicrogyria/schizencephaly, heterotopia, and cortical dysplasia). Partial correlations between diffusion changes and clinical parameters (epilepsy duration and age at disease onset) were also queried.

Results

There were significant reductions of FA, accompanied by increases in MD and λ⊥ in all segments of the CC in the patients group with both analytical methods. The absolute differences in FA were greater on ROI-analyses. There were no significant differences between the MCD subgroups, and no correlations between clinical parameters of epilepsy and FA.

Conclusions

Our study indicates DTI abnormalities consistent with microstructural changes in the corpus callosum of MCD patients. The findings support the idea that patients with epilepsy secondary to cortical malformations present widespread WM changes that extend beyond the macroscopic MRI-visible lesions.

Introduction

Malformations of cortical development (MCD) result from defects in the complex process of brain corticogenesis during early intrauterine development and are important causes of severe epilepsy and neurodevelopmental delay (Barkovich et al., 2005, Raybaud and Widjaja, 2011). Magnetic resonance imaging (MRI) is of paramount importance in the evaluation of patients with epilepsy and MCD, such as polymicrogyria, heterotopia and cortical dysplasia. Because of its excellent contrast between gray matter (GM) and white matter (WM) and high spatial resolution, MRI yields accurate in vivo diagnosis of macroscopic MCD and enables optimal topographic characterization of lesions. It may also provide prognostic factors and help in treatment planning of surgical candidates (Chang et al., 2011, Colombo et al., 2009, Diehl et al., 2010).

The role of MRI has been extended in recent years to aid in understanding of the pathogenesis of epileptic disorders as it can provide noninvasive assessment of multiple functional, metabolic and microstructural characteristics of brain tissues (Pan et al., 2008). Diffusion tensor imaging (DTI) allows parcellation of white matter tracts (DTI tractography) and can provide further information on white matter microstructure and connectivity that is not obtained with conventional MRI (Tournier et al., 2011). Most DTI studies on epilepsy have focused on patients with temporal lobe epilepsy (TLE). Several abnormalities of diffusion properties have been shown ipsilateral and contralateral to seizure onset (Arfanakis et al., 2002, Concha et al., 2009, Gross et al., 2006, Liu et al., 2012, Thivard et al., 2005). Moreover, some WM changes may persist after surgical treatment (Concha et al., 2007). Widespread white matter changes have also been demonstrated in idiopathic generalized syndromes (Li et al., 2010), especially in juvenile myoclonic epilepsy (Liu et al., 2011, O’Muircheartaigh et al., 2011). However, very few studies have devoted attention to epilepsy secondary to cortical malformations.

Quantitative abnormalities of DTI parameters have been reported in WM subjacent to MCD, as well as changes in some tracts in comparison to the contralateral side (Widjaja et al., 2007, Widjaja et al., 2009). However, the contralateral normal appearing side may not serve as an adequate control for comparisons because structural abnormalities may not be restricted to the ipsilateral hemisphere, as demonstrated with other MRI techniques, such as proton spectroscopy (Leite et al., 2007), nuclear medicine tests and electroencephalography (Iannetti et al., 1996). A comparison of MCD patients versus normal controls with voxel-based statistical parametric mapping demonstrated areas with reduced anisotropy and increased diffusivity outside the visible lesions (Eriksson et al., 2001). Voxel-wise approach offers global analyses of brain parenchyma, but it may be problematic in patients with structural malformations because it requires smoothing, alignment and registration of brain volumes to a common space (Smith et al., 2006). Tractography, on the other hand, may provide valuable information about more specific white matter tracts. Nevertheless, most of the previous tractography-based studies on MCD were seriously limited by small samples sizes (Eriksson et al., 2002, Lim et al., 2005, Munakata et al., 2006).

Herein, we hypothesized that the corpus callosum (CC), a major white matter inter-hemispheric bundle, would show microstructural changes as detected with DTI, or, more specifically, decreased FA values. The CC was chosen for analysis for the following reasons: it has demonstrated diffusion abnormalities in other epileptic syndromes (Arfanakis et al., 2002, Concha et al., 2007, Concha et al., 2009, Gross et al., 2006, Liu et al., 2012, Thivard et al., 2005), it is well identified using deterministic tractography, and its boundaries are easily depicted on sagittal slices for ROI-analyses. Unlike other bilateral WM tracts, the unpaired position of the CC in the midline makes it particularly suitable to evaluate patients with lesions in different locations in the brain, regardless of laterality. Moreover, the CC plays an important role in integration of sensory, cognitive, executive and motor functions, and it is a central component of brain connectivity networks (Huang et al., 2005, van der Knaap and van der Ham, 2011).

Our aim was to investigate quantitative diffusion characteristics in three segments of the CC in a cohort of patients with MCD (n = 32) and matched controls (n = 32). Correlations between diffusion and clinical parameters (epilepsy duration and age at disease onset) were also studied.