Evaluation of activity-based techniques to identify transient arousal in respiratory sleep disorders

Daytime sleepiness and impaired cognitive function can be a consequence of recurrent transient arousal from sleep. Arousal is often associated with abrupt changes in the electroencephalogram (EEG), and such changes can be used as an index of sleep disturbance, but EEG analysis is laborious and requires trained observers. Possible alternative indices of arousal not reliant on EEG analysis were investigated. Recordings were made from 36 sleeping subjects who were being investigated for sleep-related breathing disorders. In each study awakenings and transient arousals according to EEG criteria were compared with activity in five potential indirect indicators of arousal: wrist movement, ankle movement, left and right tibial electromyogram, and phase change in ribcage-abdominal movement. The mean values of sensitivity to arousal ranged from only 25 to 45%. However, their high positive predictive accuracies (PPAs, 68 to 92%) indicated that activity, when present, was usually associated with arousal. Sensitivity to awakenings was higher (71-87%), though PPAs were lower (42 to 63%). For the indicator based on ribcage-abdominal phase, the number of periods of activity showed a significant relation to the number of arousals (r = 0.70, P < 0.001). It can be concluded that phase changes in chest/abdomen movement are a useful indicator of arousal associated with obstructive apnoea and related conditions. Limb activity has much lower sensitivity for transient arousal, but may be of value in indicating periods of wakefulness.