Twenty-four-hour recording in REM-narcoleptics with special reference to nocturnal sleep disruption.

Twenty narcoleptic patients and ten age-matched normals were polygraphically monitored for 58 consecutive hours. All subjects were on regimented sleep (hours between 2230 and 0700). Group A (11 patients and 10 normals) had enforced wakefulness during the day whereas Group B (9 patients) were permitted to sleep (mean = 2 1/2 hr.). On day 2, all subjects were permitted to sleep for 15-min periods every 2 hr. In narcoloptics, sleep recordings demonstrated a reduction of sleep latency, an increase of stage 1, and a decrease in stages 3 and 4 compared to normals, but total REM time and percentage of REM sleep were similar. Groups A and B showed no difference in the incidence of nocturnal awakenings. REM cyclic periodicity was larger in narcoleptics who also demonstrated a REM-sleep fragmentation. This fragmentation became more pronounced as time passed, with several shifts from REM to wakefulness and stage 1. Narcoleptics present REM onset sleep period but also show an inability to remain in REM sleep.