Functional MR imaging of language, memory, and sensorimotor cortex☆

This study aimed to evaluate lexical retrieval, presurgery and postsurgery, among children and adolescents who had undergone temporal lobe resection for intractable epilepsy and to compare outcomes in patients whose surgery involved the left temporal lobe or the right temporal lobe.

A retrospective chart review identified 36 patients from a major pediatric epilepsy treatment center who had undergone temporal lobe resection (21 underwent left temporal lobe resection; 15 underwent right temporal lobe resection) for intractable epilepsy and who had completed neuropsychological testing that included a measure of confrontation naming (Boston Naming Test, BNT) and verbal fluency (Delis–Kaplan Executive Function System (D-KEFS) Fluency) prior to and after surgery. Linear mixed effects regression models were used to evaluate presurgery and postsurgery changes and to compare the left temporal lobe resection group with the right temporal lobe resection group.

Confrontation naming performance declined after left, but not right, temporal lobe resection (p < 0.05). This effect was not documented for verbal fluency.

Left temporal lobe resection for intractable epilepsy is associated with a decline in lexical retrieval. The risk of decline in specific language functions following surgery involving the left temporal lobe should be incorporated in the counseling of patients and families in decision-making with regard to surgery.

Pre-operative tractography of the optic radiation (OR) has been advised to assess the risk for postoperative visual field deficit (VFD) in certain candidates for resective epilepsy surgery. Diffusion tensor imaging (DTI) tractography relies on a precise anatomical determination of start and target regions of interest (ROIs), such as the lateral geniculate nucleus (LGN) and the primary visual cortex (V1). The post-chiasmal visual pathway and V1 show considerable inter-individual variability, and in epilepsy patients parenchymatous lesions might further complicate this matter. A functionally based tractography (FBT) seems beneficial for precise OR identification. We assessed practicability of FBT for OR identification in a patient with occipital lobe epilepsy due to a temporo-occipital maldevelopmental tumor. The MRI protocol at 3 T included a T1-weighted sagittal 3D scan, a T2-weighted axial 2D scan and a DTI scan using an echo planar spin echo sequence. ROIs for fiber tracking of OR (LGN & V1) were determined with T2*-weighted fMRI-based retinotopic assessment. After DTI pre-processing and fiber tracking, paths with similar properties were combined in clusters for visual presentation and OR localization. Retinotopic phase maps allowed for the identification of V1 and LGN for a precise DTI-based reconstruction of OR, which was distant to the patient's tumor. Location and structure of ORs were comparable in each hemisphere. FBT could thus influence the human research of the extrastriate visual pathway and the risk management of post-operative VFD in epilepsy surgery.

Blood oxygen level-dependent (BOLD) functional magnetic resonance imaging (fMRI) is used primarily to identify the location of eloquent brain functions to spare during epilepsy surgery. Motor tapping yields robust identification of the primary motor cortex, and visual checkerboard stimuli identify primary visual cortex. Language-processing areas in frontal and temporal regions are best identified with a panel of paradigms that target receptive and expressive aspects language and also provide redundancy to assure reproducibility of findings. Language fMRI predicts postoperative naming outcomes and has excellent but not complete agreement with invasive language assessments. There are clear and identifiable circumstances when fMRI may provide false lateralizing or localizing information. Recent memory-encoding paradigms demonstrate material specificity of hippocampal function. Functional adequacy rather than functional reserve predict postoperative memory outcomes. Diffusion tensor imaging (DTI) tractography may be melded with fMRI to identify the long white-matter tracts that may be at risk in epilepsy surgery. Interictal spike fMRI, in some circumstances, may be able to identify the epileptogenic zone, but has not yet entered standard clinical practice.

For an understanding of cognitive impairment in epilepsy, it is essential to appreciate that static and dynamic factors affect brain function in this disease. Whereas morphological lesions or structural changes are associated with more or less irreversible deficits, epileptic activity, seizures, and the treatment of epilepsy can cause dynamic and largely reversible impairments. The contribution of these factors varies depending on the type of epilepsy, the age at lesion/epilepsy onset, the localization and lateralization of the epilepsy, and individual demographic patient characteristics. Altered brain structure and function can result in epilepsy, but epilepsy can also alter the functional cerebral organization of the brain. Thus, epilepsy-related cognitive impairment must be seen within a developmental neuropsychological framework. From a neuropsychological point of view, it is essential as to whether epilepsy affects the maturing versus the mature or aging brain. Epilepsy can result in retardation, loss of acquired functions, or accelerated mental decline. Cognitive impairments in epilepsy often exist from the commencement of the epilepsy; early-onset lesions/epilepsy interfere with mental development; and a progressive etiology, severe seizures, and lesions secondary to epilepsy can accelerate mental decline. Uncontrolled epilepsy and epileptic activity, as well as the treatment of epilepsy, may reversibly and irreversibly affect cognition. Within this framework, neuropsychology has become an essential diagnostic tool for the early detection and monitoring of cognitive impairment and its determinants in newly established and chronic epilepsies. Neuropsychology serves as a valuable tool for quality and outcome control of the treatment of epilepsy and helps to improve the individual medical care of patients with epilepsy.

In patients with left temporal lobe epilepsy (TLE) due to hippocampal sclerosis (HS) decreased naming ability is common, suggesting a critical role for the medial left temporal lobe in this task. We investigated the integrity of language networks with functional MRI (fMRI) in controls and TLE patients.

We performed an fMRI verbal fluency paradigm in 22 controls and 66 patients with unilateral mesial TLE (37 left HS, 29 right HS). Verbal fluency and naming ability were investigated as part of the standard presurgical neuropsychological assessment. Naming ability was assessed using a visual confrontation naming test.

Left TLE patients had significantly lower naming scores than controls and those with right TLE. Right TLE patients performed less well than controls, but better than those with left TLE. Left TLE had significantly lower scores for verbal fluency than controls.

In controls and right TLE, left hippocampal activation during the verbal fluency task was significantly correlated with naming, characterised by higher scores in subjects with greater hippocampal fMRI activation. In left TLE no correlation with naming scores was seen in the left hippocampus, but there was a significant correlation in the left middle and inferior frontal gyri, not observed in controls and right TLE. In left and right TLE, out of scanner verbal fluency scores significantly correlated with fMRI activation for verbal fluency in the left middle and inferior frontal gyri.

Good confrontation naming ability depends on the integrity of the hippocampus and the connecting fronto-temporal networks. Functional MRI activation in the left hippocampus during verbal fluency is associated with naming function in healthy controls and patients with right TLE. In left TLE, there was evidence of involvement of the left frontal lobe when naming was more proficient, most likely reflecting a compensatory response due to the ongoing epileptic activity and/or underlying pathology.