Abstracts

Rationale: Many patients with epilepsy experience cognitive deficits that adversely affect their quality of life. It is well established that sleep plays an important role in the consolidation of recently formed memories, with relative improvement in memory performance after a night of sleep compared to an equivalent period of daytime wakefulness. Most of the previous studies elucidating these findings were performed in healthy individuals. Whether and how this critical memory processing is altered in epilepsy is unknown. This pilot study assessed patients with epilepsy for the presence of memory consolidation, and the possible impact of seizures.Methods: Patients with epilepsy were recruited prospectively from the adult epilepsy monitoring unit at Brigham and Women’s Hospital on day 2 of their admission. Inclusion criteria included age (18-60 years) and a diagnosis of focal-onset seizures. Patients with a prior craniotomy, barbiturates/benzodiazepines use, inability to achieve >40% retention rate on the memory task, or substance abuse were excluded. The memory task consisted of 15 pairs of colored pictures showing different animals and high frequency objects on a 5 x 6 grid. Subjects were repeatedly shown the picture-pair locations until they were able to recall at least 40% of the items (training session). An average of 12 hours later, subjects were tested on the material they trained on, and then they were trained on another set of pictures. Subjects were trained and tested every 12 hours until they had a seizure, were sleep deprived, or after 4 days if no seizure occurred. EMG chin electrodes were placed to enable sleep scoring. In addition, the number and density of sleep spindles were assessed using a wavelet-based algorithm.Results: A total of 11 subjects were analyzed (Table 1). Three subjects had seizures during the day, another 3 had nocturnal seizures. The percentage of memory retention over 12 hours of wakefulness was 62.7% % and over 12 hours that included sleep was 84.6 % (p = 0.04). Average time in stage I sleep was 14%, stage II 63%, REM 15%, and slow wave sleep 8%. Total spindle counts ranged between 170 to 506, and spindle density during stage II sleep ranged between 0.77 to 1.88 spindles per minute. Performance on overnight testing correlated best with the duration of slow wave sleep in minutes (r = 0.62, p= 0.049), but did not correlate with total number of sleep spindles or spindle density. The 3 subjects who had seizures during the day did not have a change in memory retention rates, while the 3 subjects who had nocturnal seizures had a drop in retention from an average of 92% to 60.5%.Conclusions: Adult epilepsy patients consolidate memories in the EMU. This sleep dependent memory consolidation is most closely correlated with the amount of time spent in slow wave sleep. Preliminary data show those patients with nocturnal seizures had worse performance on the memory task suggesting that seizures may disrupt sleep-dependent memory consolidation. Larger prospective studies are needed to confirm these findings.