The arousal and breathing responses to total airway occlusion during sleep were measured in 12 normal subjects (7 males and 5 females) aged 25-36 yr. Subjects slept while breathing through a specially designed nosemask, which was glued to the nose with medical-grade silicon rubber. The lips were sealed together with a thin layer of Silastic. The nosemask was attached to a wide-bore (20 mm ID) rigid tube to allow a constant-bias flow of room air from a blower. Total airway occlusion was achieved by simultaneously inflating two rubber balloons fixed in the inspiratory and expiratory pipes. A total of 39 tests were done in stage III/IV nonrapid-eye movement (NREM) sleep in 11 subjects and 10 tests in rapid-eye-movement (REM) sleep in 5 subjects. The duration of total occlusion tolerated before arousal from NREM sleep varied widely (range 0.9-67.0 s) with a mean duration of 20.4 +/- 2.3 (SE) s. The breathing response to occlusion in NREM sleep was characterised by a breath-by-breath progressive increase in suction pressure achieved by an increase in the rate of inspiratory pressure generation during inspiration. In contrast, during REM sleep, arousal invariably occurred after a short duration of airway occlusion (mean duration 6.2 +/- 1.2 s, maximum duration 11.8 s), and the occlusion induced a rapid shallow breathing pattern. Our results indicate that total nasal occlusion during sleep causes arousal with the response during REM sleep being more predictable and with a generally shorter latency than that in NREM sleep.
Coin Flipping in Dynamic Programming Is Almost Useless
