Ambulatory circadian monitoring (ACM) based on thermometry, motor activity and body position (TAP): a comparison with polysomnography.

An integrated variable based on the combination of wrist Temperature, motor Activity and body Position (TAP) was previously developed at our laboratory to evaluate the functioning of the circadian system and sleep-wake rhythm under ambulatory conditions. However, the reliability of TAP needed to be validated with polysomnography (PSG). 22 subjects suffering from sleep disorders were monitored for one night with a temperature sensor (iButton), an actimeter (HOBO) and exploratory PSG. Mean waveforms, sensitivity (SE), specificity (SP), agreement rates (AR) and comparisons between TAP and sleep stages were studied. The TAP variable was optimized for SE, SP and AR with respect to each individual variable (SE: 92%; SP: 78%; AR: 86%). These results improved upon estimates previously published for actigraphy. Furthermore, TAP values tended to decrease as sleep depth increased, reaching the lowest point at phase 3. Finally, TAP estimates for sleep latency (SL: 37±9 min), total sleep time (TST: 367±13 min), sleep efficiency (SE: 86.8±1.9%) and number of awakenings (NA>5 min: 3.3±.4) were not significantly different from those obtained with PSG (SL: 29±4 min; SE: 89.9±1.8%; NA>5 min: 2.3±.4), despite the heterogeneity of the sleep pathologies monitored. The TAP variable is a novel measurement for evaluating circadian system status and sleep-wake rhythms with a level of reliability better to that of actigraphy. Furthermore, it allows the evaluation of a patient's sleep-wake rhythm in his/her normal home environment, and at a much lower cost than PSG. Future studies in specific pathologies would verify the relevance of TAP in those conditions.