Whole-brain voxel-based statistical analysis of gray matter and white matter in temporal lobe epilepsy
Introduction
In clinical practice, the investigation and treatment of patients with epilepsy have been revolutionized by the advent of MRI, which has been demonstrated to be a reliable and accurate indicator of pathologic findings underlying epilepsy. Techniques such as volumetric acquisition with thin contiguous slices and three-dimensional reformatting have enhanced the ability of MRI to display brain anatomy and to visualize epileptogenic brain lesions in vivo.
In temporal lobe epilepsy (TLE), most MRI studies have focused on the assessment of the hippocampus based on pathological Babb and Brown, 1987, Falconer et al., 1964, Mouritzen Dam, 1980 and electrophysiological (Quesney, 1986) evidence from early studies for its involvement in the epileptogenic process. More recent MRI studies have shown that structural brain abnormalities associated with TLE extend beyond the hippocampus, involving other mesial and limbic structures such as the entorhinal cortex Bernasconi et al., 1999, Bernasconi et al., 2003a, Bernasconi et al., 2003b, Salmenpera et al., 2000, the thalamus DeCarli et al., 1998, Natsume et al., 2003, and the fornix Baldwin et al., 1994, Kuzniecky et al., 1999. There is also MRI evidence for reduction in the total temporal neocortical gray matter and white matter volumes Coste et al., 2002, Jutila et al., 2001, Lee et al., 1998, Moran et al., 2001. The effect of mesolimbic pathology on the rest of the brain has not been thoroughly evaluated.
To date, most MRI morphometric studies have been based on manual delineation of single brain structures. This procedure, which is labor intensive, suffers from difficulties related to defining reliable anatomical boundaries and may therefore result in low intra- and interrater reliability. These difficulties explain in part why only a selected number of structures have been evaluated so far. Voxel-based morphometry (VBM) is a fully automated technique allowing identification of regional differences in the amount of gray matter (GM) and white matter (WM) with no a priori region of interest, enabling an objective analysis of the whole brain between groups of subjects (Ashburner and Friston, 2000). This technique has been used to reveal pathological changes in GM and WM in various neurological conditions, including Alzheimer's disease and schizophrenia Karas et al., 2003, Kubicki et al., 2002. Voxel-based morphometry of the GM in TLE has produced conflicting results. In one study, no GM differences were found between TLE patients and healthy controls (Woermann et al., 1999). In another study, areas of decreased GM were found in the hippocampus ipsilateral to the seizure focus and in extratemporal areas (Keller et al., 2001). VBM has shown subtle differences in frontal lobe GM in relation to aggression in TLE (Woermann et al., 2000). None of these studies examined WM changes associated with TLE.
The purpose of this study was to examine whole-brain GM and WM changes in TLE using VBM and investigate the relationship between these changes and clinical parameters such as febrile convulsions, age of onset, duration of epilepsy, and secondary generalized seizures.
Section snippets
Methods
We selected 85 consecutive patients with medically intractable TLE (44 males, mean age ± SD = 35 ± 10 years, range = 16–54) and unilateral hippocampal atrophy as determined by volumetric MRI (see below). Patients were compared to 47 neurologically normal controls (24 males, mean age = 33 ± 12 years, range = 20–66). The patients with right TLE and left TLE and healthy controls did not differ in age [ANOVA; F(2,125) = 0.88, P = 0.4] or sex distribution (for right TLE: χ2 = 2.6, df = 1, P = 0.1;
Areas of GM reduction in TLE patients compared to healthy controls (see Table 1A)
Areas of decreased GM were observed in the temporolimbic and frontal areas (Figs. 1A and B) (Table 1A).
Discussion
Voxel-based morphometry showed evidence for both limbic and extralimbic pathology in patients with pharmacologically intractable TLE.
Acknowledgements
This research was funded by a grant from the Canadian Institutes of Health research (CIHR-12483).
References (51)
- et al.
Voxel-based morphometry—The methods
NeuroImage
(2000) - et al.
Automatic differentiation of anatomical patterns in the human brain: validation with studies of degenerative dementias
NeuroImage
(2002) - et al.
A comprehensive study of gray matter loss in patients with Alzheimer's disease using optimized voxel-based morphometry
NeuroImage
(2003) - et al.
Voxel based morphometry of hippocampal and extra-hippocampal effects of unilateral temporal lobe epilepsy
NeuroImage
(2001) - et al.
Voxel-based morphometric analysis of gray matter in first episode schizophrenia
NeuroImage
(2002) - et al.
Quantitative MRI volumetry of the entorhinal cortex in temporal lobe epilepsy
Seizure
(2000) - et al.
Memory: brain systems and behavior
Trends Neurosci
(1988) - et al.
Some connections of the entorhinal (area 28) and perirhinal (area 35) of the rhesus monkey. II. Frontal lobe afferents
Brain Res
(1975) - et al.
Voxel-by-voxel comparison of automatically segmented cerebral gray matter—A rater-independent comparison of structural MRI in patients with epilepsy
NeuroImage
(1999) - et al.
Pathological findings in epilepsy