Primary or working memory in frontal lobe epilepsy: An 18FDG-PET study of dysfunctional zones

We previously demonstrated that patients with frontal lobe epilepsy show deficits on a visual working memory paradigm and that this paradigm produces increased 18FDG uptake in the dorsolateral prefrontal cortex (DPFC), premotor cortex, angular and supramarginal gyri, basal forebrain, and ventral frontal poles of normal subjects when compared with a control task. We hypothesized that subjects with frontal lobe epilepsy would have impaired frontal activation during this task.

One resting and two activated images were obtained with 18FDG-PET in 15 subjects and 14 controls. One was a delayed (DMS) and one an immediate (IMS) match to sample paradigm. Discriminant and factor analyses were used to analyze the data, supplemented by selected t tests.

No differences in glucose uptake were found between the DMS and IMS in the epilepsy subjects, in distinct contrast to controls. A comparison between controls and epilepsy subjects showed differences both ipsilateral and contralateral to the epileptic focus in the frontal regions involved in the task, with small changes in nonfrontal, task-related regions as well. The task itself brought out or highly exaggerated differences seen at rest. There was weak evidence that other frontal and temporal regions were attempting to compensate for the DPFC deficit.

A unilateral epileptic focus is capable of suppressing function along a large task-related circuit ipsilateral and contralateral to the focus. Peripheral cortical regions compensate poorly for the area of dysfunction.