Objective: Temporal lobe epilepsy (TLE) affects brain networks and is associated with impairment of episodic memory. Temporal and extratemporal reorganization of memory functions is described in functional magnetic resonance imaging (fMRI) studies. Functional reorganizations have been shown at the local activation level, but network-level alterations have been underinvestigated. We aim to investigate the functional anatomy of memory networks using memory fMRI and determine how this relates to memory function in TLE.

Perampanel, a noncompetitive antagonist of the postsynaptic a‐amino‐3‐hydroxy‐5‐methyl‐4‐isoxazolepropionic (AMPA) receptor, is effective for controlling focal to bilateral tonic–clonic seizures but is also known to increase feelings of anger. Using statistical parametric mapping–derived measures of activation and task‐modulated functional connectivity (psychophysiologic interaction), we investigated 14 people with focal epilepsy who had verbal fluency functional magnetic resonance imaging (fMRI) twice, before and after the add‐on treatment of perampanel. For comparison, we included 28 people with epilepsy, propensity‐matched for clinical characteristics, who had two scans but no change in anti‐seizure medication (ASM) regimen in‐between. After commencing perampanel, individuals had higher task‐related activations in left orbitofrontal cortex (OFC), fewer task‐related activations in the subcortical regions including the left thalamus and left caudate, and lower task‐related thalamocaudate and caudate‐subtantial nigra connectivity. Decreased task‐related connectivity is observed between the left OFC and precuneus and left medial frontal lobe. Our results highlight the brain regions associated with the beneficiary therapeutic effects on focal to bilateral tonic–clonic seizures (thalamus and caudate) but also the undesired affective side effects of perampanel with increased anger and aggression (OFC).

Episodic memory functions are impaired in up to 40% of patients with temporal lobe epilepsy (TLE) following anterior temporal lobe resection (ATLR). Little is known of plasticity effects beyond one year postoperatively. We describe the dynamic plasticity of memory encoding networks long-term after ATLR.Sixteen patients with TLE (8 left-sided) and six controls performed a memory-encoding fMRI paradigm of faces and words across equivalent time-points (pre-surgery, 4, 12 months, and 7–10 years post-surgery). Flexible factorial designs probed the changes in fMRI activations in patients between 12 months and long-term postoperatively, relative to longitudinal changes in controls. Clinical memory scores (list- and design-learning) were correlated with changes in activations.From 1 to 10 years postoperatively, prefrontal cortex (PFC) activations were increased ipsilaterally for word encoding after left ATLR and bilaterally for face encoding after left and right ATLR. After left ATLR, improvement in visual memory correlated with increased contralateral amygdala, insula, and PFC acti- vations. After right ATLR, right fusiform gyrus and temporo-parietal activations positively correlated with visual memory. Better verbal memory correlated with increased right fusiform gyrus and PFC activations.Years after left and right ATLR, there is continued contralateral and ipsilateral plasticity which is supportive of memory functions.

No abstract