MRI analysis in temporal lobe epilepsy: cortical thinning and white matter disruptions are related to side of seizure onset

Abstract

Purpose: Past studies reported more widespread structural brain abnormalities in patients with left compared to right temporal lobe epilepsy (TLE), but the profile of these differences remains unknown. This study investigated the relationship between cortical thinning, white matter compromise, epilepsy variables, and the side of seizure onset, in patients with TLE.

Methods: We performed diffusion tensor imaging tractography and cortical thickness analyses of 18 patients with left TLE (LTLE), 18 patients with right TLE (RTLE), and 36 controls. We investigated the relationship among brain structural abnormalities, side of seizure onset, age of seizure onset, and disease duration.

Key findings: Patients with TLE displayed cortical thinning and white matter compromise, predominately on the side ipsilateral to the seizure onset. Relative to RTLE, patients with LTLE showed more widespread abnormalities, particularly in white matter fiber tracts. Greater compromise in white matter integrity was associated with earlier age of seizure onset, whereas cortical thinning was marginally associated with disease duration.

Significance: These data support previous findings of LTLE showing greater structural compromise than RTLE, and suggest that mechanisms may not be uniform for gray and white matter compromise in patients with LTLE and RTLE. These results may indicate that LTLE is different from RTLE, possibly due to greater vulnerability of the left hemisphere to early injury and the progressive effects of seizures.

Figure 1

Sagittal views of the selected fiber tracts. Individual fiber tracts are shown projected on their corresponding T1-weighted images using Tractoview software. Color-coding is included to assist with identification of the fibers in the superimposed images.

Figure 2

White matter fractional anistropy (FA) of selected fiber tract ROIs in patients with LTLE and RTLE and healthy controls. Error bars represent standard errors.

Figure 3

Cluster-based t-statistic surface maps of cortical thinning in patients with LTLE versus controls (top), RTLE versus controls (middle), and RTLEs versus LTLEs (bottom). Cluster maps are thresholded at t > 2, corrected p-values < .05. In the top and middle graphs, blue areas represent areas of thinner cortex in patients compared to controls and red areas represent areas of thinner cortex in controls compared to patients. In the bottom graph red areas represent thinner cortex in patients with LTLE compared to patients with RTLE. Areas of blue represent thinner cortex in patients with RTLE.