Quantitative MR and cognitive impairment in cryptogenic localisation-related epilepsy.

For patients with chronic cryptogenic localisation-related epilepsy (CLRE), conventional MRI does not provide measures to discern between patients with or without cognitive complaints. We investigated, in a preliminary study, whether it is possible to detect cerebral biomarkers of cognitive impairment in patients with CLRE using sensitive quantitative MRI techniques. Neuropsychological assessment and quantitative 3.0 T MRI, comprising T2 relaxometry, diffusion tensor imaging, and spectroscopic imaging, were applied to 35 patients with CLRE and 21 healthy controls. Analysis included the left and right hippocampi, and frontal and temporal lobes. Differences between the groups and correlations with cognitive and clinical characteristics were assessed. Patients with epilepsy scored significantly worse on cognitive tasks compared to healthy controls. Significantly larger CSF fractions in the hippocampi and left temporal lobe, a longer T2 relaxation time in the left hippocampus, and a significantly higher concentration of glutamate/glutamine in the left frontal lobe were observed in patients with epilepsy. Moreover, poor memory performance was significantly correlated with larger CSF fractions in the right hippocampus and left temporal lobe in patients. In the temporal lobe, an association between subtle changes in morphology (indicative of atrophy) and memory performance was found, consistent with previous literature. These results may help to explain the alterations in brain functioning in patients with epilepsy.

Functional MRI in chronic epilepsy: associations with cognitive impairment.

Chronic epilepsy is frequently accompanied by serious cognitive side-effects. Clinical factors are important, but cannot account entirely for this cognitive comorbidity. Therefore, research is focusing on the underlying cerebral mechanisms to understand the development of cognitive dysfunction. In the past two decades, functional MRI techniques have been applied extensively to the study of cognitive impairment in chronic epilepsy. However, because of wide variation in study designs, analysis methods, and data presentation, interpretation of these studies has become increasingly difficult for clinicians. In patients with localisation-related epilepsy, whether findings of functional MRI represent the underlying neuronal substrate for cognitive decline remains a subject of debate.

White matter lesions in patients with localization-related epilepsy.

OBJECTIVES: White matter lesions (WML) have been proven to be associated with cognitive impairment. As (1) the decline of cognitive function is the most frequent comorbid disorder in epilepsy, and (2) patients with epilepsy have a relatively high prevalence of WML, the question is raised whether WML in patients with epilepsy are also associated with cognitive decline. MATERIALS AND METHODS: A high-resolution magnetic resonance imaging examination was performed at 3.0 T, comprising T1-weighted, T2 relaxometry, and fluid-attenuated inversion recovery (FLAIR) sequences. Patients with localization-related epilepsy with impaired and unimpaired cognitive functioning and a healthy control group were included. Furthermore, the performance of an automated WML detection algorithm, based on regional intensity evaluation, was assessed. RESULTS: The prevalence of WML, detected on 3.0 T FLAIR images, is 63% in healthy volunteers and 46% in patients with localization-related, cryptogenic epilepsy. No relationship between WML volume and cognitive performance was observed. The WML volumes from the automated segmentation method were found to be significantly correlated to the volumes obtained by neuroradiologic assessment. CONCLUSIONS: No relations could be found between WML and cognition in the well-defined population of patients with epilepsy. Other clinical characteristics of chronic epilepsy, such as seizures, age of onset, and medication are more likely to play an important role in cognitive decline. Furthermore, the automated WML detection algorithm using a regional Z-score analysis can successfully segment and quantify the WML on FLAIR images.