The impact of sleep characteristics and epilepsy variables on memory performance in patients with focal seizures

Frank J van Schalkwijk,Monica Ricci,Armin Nikpour,Laurie A Miller
DOI: https://doi.org/10.1016/j.yebeh.2018.06.034
Abstract:Disturbed sleep can negatively affect overnight memory retention as well as new learning the subsequent day. In healthy participants, positive associations between memory performance and sleep characteristics (e.g., time spent in slow-wave sleep [SWS]) have been detected. In a previous study, we found that SWS was much reduced in patients with focal seizures, but when correlations between memory complaints and various sleep characteristics were considered, the only significant relationship was with the time to onset of rapid eye movement (REM) sleep (i.e., REM latency). In this study, we investigated the relationships between sleep, epilepsy, and objective memory performance variables. Twenty-five patients with focal seizures had their memory tested while undergoing a two-day ambulatory electroencephalography (EEG). The sleep variables of interest were the percentage of time spent in SWS (%SWS) and REM latency. Epilepsy variables included the presence of (1) seizures, (2) interictal epileptiform discharges, and/or (3) hippocampal lesions as well as site of seizure origin (temporal vs extratemporal). Overnight retention (of autobiographical events, a story, and a complex geometric figure) and the ability to learn a word list on day 2 were the measures of memory. A significant positive correlation was found between word-list learning and %SWS during the previous night. A significant negative correlation was observed between REM latency and overnight retention of autobiographical events. Overnight retention scores for the story and geometric figure were not related to sleep characteristics but were negatively affected by the presence of epileptiform activity. Story retention was also worse for temporal lobe epilepsy (TLE) than for patients with extratemporal epilepsy (ETE). Those with hippocampal lesions were more impaired than those without lesions on word-list learning, autobiographical events' retention, and story retention. When multiple contributing factors were entered into regression analyses, %SWS was found to be the best predictor of subsequent word-list learning, whereas the presence of a hippocampal lesion was the best predictor of overnight retention of autobiographical events and a story. These findings provide further evidence of the ways in which particular sleep characteristics are associated with memory and suggest that treatment of sleep disturbances in patients with epilepsy might be helpful for improving their performance.
What problem does this paper attempt to address?