Sleep architecture changes during a trek from 1400 to 5000 m in the Nepal Himalaya

The aim of this study was to examine sleep architecture at high altitude and its relationship to periodic breathing during incremental increases in altitude. Nineteen normal, sea level-dwelling volunteers were studied at sea level and five altitudes in the Nepal Himalaya. Morning arterial blood gases and overnight polysomnography were performed in 14 subjects at altitudes: 0, 1400, 3500, 3900, 4200 and 5000 m above sea level. Subjects became progressively more hypoxic, hypocapnic and alkalinic with increasing altitude. As expected, sleep architecture was affected by increasing altitude. While time spent in Stage 1 non-rapid eye movement sleep increased at 3500 m and higher (P < 0.001), time spent in slow-wave sleep (SWS) decreased as altitude increased. Time spent in rapid eye movement (REM) sleep was well preserved. In subjects who developed periodic breathing during sleep at one or more altitudes (16 of 19), arousals because of periodic breathing predominated, contributing to an increase in the total arousal index. However, there were no differences in sleep architecture or sleeping oxyhaemoglobin saturation between subjects who developed periodic breathing and those who did not. As altitude increased, sleep architecture became progressively more disturbed, with Stage 1 and SWS being affected from 3500 m, while REM sleep was well preserved. Periodic breathing was commonplace at all altitudes, and while associated with increases in arousal indices, did not have any apparent effect on sleep architecture.