Abstracts

RATIONALE: Psychogenic nonepileptic seizures (PNES) are a significant public health problem, representing about 25% of patients admitted to epilepsy monitoring units. The most accurate test to verify the presence of PNES is video-EEG monitoring with recording of a typical spell, however this is not possible in all cases. Other clinical characteristics which distinguish these patients from those with epilepsy would be helpful from both a clinical and scientific standpoint. We looked at sleep structure in patients with PNES and compared this to patients with epileptic seizures.
METHODS: Patients admitted consecutively for diagnosis or surgical evaluation were included. Epileptic (ES) and nonepileptic seizures were verified by video-EEG monitoring. No complex partial or secondarily generalized seizures occurred within 24 hours of polysomnography or during recording. Patients were on no caffeine, anticonvulsant or antidepressant medication, and patients with known or suspected sleep disorders were excluded. Subjects were not sleep deprived and were not allowed daytime naps. Polysomnography was performed by reformatting digital EEG to polysomnographic parameters. REM latency was measured as time from sleep onset until the start of the first REM period. Comparisons were made with students t-test, with p.[lt] .05 considered significant. All patients with PNES were evaluated by a psychiatrist
RESULTS: Five patients with PNES had a total of ten polysomnograms (2 per patient); all were women. Five female ES patients in the same age group (less than 60) had at least two recordings; the first two recordings in each patient were used in the study. The average age of PNES patients was 31 (range 18-47); of epilepsy patients the average age was 33 (range 27-45). Percent REM sleep was significantly greater in patients with PNES (24 +/- 1%) compared to patients with epilepsy (20+/-2%). There were no significant differences in slow wave sleep or sleep efficiency. REM latency was slightly less in PNES patients although this difference was not statistically significant. Additionally, there were no differences for PNES vs. epileptic seizure patients in Stage 1 (6.6 +/- 1.4% vs. 6.7 +/- 0.9%), stage 2 (54.6 +/- 2.1% vs. 47.5 +/- 1.8%), or total sleep time (443 +/- 22 min vs. 470 +/- 16 min).
CONCLUSIONS: The results show increased REM in patients with PNES compared with ES. All other parameters studied are similar. This could represent decreased REM in ES or increased REM in PNES. The study is important for three reasons. First, it suggests that patients with PNES may suffer from major depression, which is known to be associated with increased REM sleep, even when not apparent on questioning. This finding would need to be confirmed in a larger group of patients, preferably using standardized depression inventories. Secondly, the study suggests that sleep patterns could help in distinguishing PNES from ES. At present this is not practical, but could be as automated staging improves. Perhaps in combination with other criteria such as normal EEG and MRI, it might be possible to identify patients with a high likelihood of having PNES when ictal recordings cannot be obtained. Finally, it further implicates the role of sleep (particularly REM) in psychiatric illness. This poorly understood area, although interesting, awaits further research.