Validation of a Device for the Ambulatory Monitoring of Sleep Patterns: A Pilot Study on Parkinson's Disease

Abstract

The development of wearable devices has increase interest in the use of ambulatory methods to detect sleep disorders more objectively than those permitted by subjective scales evaluating sleep quality, while subjects maintain their usual lifestyle. This study aims to validate an ambulatory circadian monitoring (ACM) device for the detection of sleep and wake states and apply it to the evaluation of sleep quality in patients with Parkinson disease (PD). A polysomnographic validation study was conducted on a group of patients with different sleep disorders in a preliminary phase, followed by a pilot study to apply this methodology to PD patients. The ACM device makes it possible to estimate the main sleep parameters very accurately, as demonstrated by: (a) the lack of significant differences between the mean values detected by PSG and ACM in time in bed (TIB), total sleep time (TST), sleep efficiency (SE), and time awake after sleep onset (WASO); (b) the slope of the correlation lines between the parameters estimated by the two procedures, very close to 1, which demonstrates the linearity of the predictions; (c) the low bias value in the estimates obtained through ACM. Sleep in PD is associated with lower distal skin temperature, efficiency and overall sleep time; greater WASO, activity during sleep and duration of naps and a worse circadian function index. In summary, the ACM device has proven to be clinically useful to evaluate sleep in an objective manner, thanks to the integrated management of different complementary variables, having advantages over conventional actigraphy.

Keywords: Parkinson's disease; actigraphy; ambulatory recordings; circadian rhythms; polysomnography; sleep; thermometry.

Figures

Figure 1

Sequence for data processing and algorithms used for automatic detection of sleep and wake periods. Downloaded raw data from ACM device (A) were submitted to a two-phase procedure. The first procedure for automatic sleep and wake (B,C) detection used the TAPL algorithm (D) (integrating wrist temperature, motor activity, variability in wrist position and variability in light exposure), implemented on the Kronowizard website (https://kronowizard.um.es/, University of Murcia). Given that, WT rhythm is the inverse of the A, P, and L, WT values were inverted before calculating the mean of the four standardized variables. Therefore, a TAPL value of 0 indicates deep rest, characterized by immobility, vasodilation of the skin vessels, and low variability of L exposure, while 1 corresponds to wake, movement, vasoconstriction, and high light variability. A concrete period was classified as sleep when TAPL values fell beneath a preset threshold (D), previously validated by PSG (11). Next, we proceeded to remark sleep episodes using the Keywake algorithm implemented on the Kronowizard website (https://kronowizard.um.es/, University of Murcia) in order to improve the precision of the estimates (E).

Figure 2

Decision trees for the marking, using criteria objectifiable by an expert, of the sleep interval defined as the time between the voluntary start of sleep (bed time) and the end of the time spent in bed (get up time). The expert uses the criteria of: the event marker (E), lights on and off (L), motor activity (A), and stability of wrist position (P).

Figure 3

Representative examples of two sleep pathologies (apnea and insomnia), for which a comparison between the hypnogram determined by PSG and the sleep pattern obtained by ACM recording is shown. The sleep detected by the ACM device is shown in orange, while the awakenings appear in white. The estimation of sleep and wake episodes were determined automatically based on the integration of sleep temperature, light exposure (visible and infrared), time in movement, and hand position. The corresponding hypnogram has been superimposed on the bottom of each panel to facilitate comparison.

Figure 4

Pearson correlations showing the correspondence between the main sleep parameters estimated by ACM and PSG. TIB, time in bed; TST, total sleep time; SE, sleep efficiency; WASO, wake time after sleep onset. The graph indicates the corresponding equation, its R value and its probability.

Figure 5

Bland-Altman representation, comparing the deviations in the estimates generated by ACM and PSG. ACM overestimates time in bed by 0.58% and underestimates total sleep time by 1.05%, sleep efficiency by 0.48%, and WASO (wake time after sleep onset) by 10.5%. Each of the graphs shows with horizontal lines the mean deviation ± 1.96 SD.

Figure 6

Weekly recording (A,C) and one-night recording (B,D) via ACM of wrist temperature (red line), exposure to visible light (blue line), acceleration (green line), time in movement (brown line), wrist position (dark green line), and estimated sleep (orange bars), representative of two subjects monitored in the study: one control subject (A,B) and a patient with PD (C,D). Surprising is the great fragmentation of sleep, accompanied by high levels of movement, frequent lights-on episodes and temperature drops of the skin during sleep that were observed in the patient with PD.

Figure 7

Mean 24-h wave of sleep probability in patients with PD (red line) and healthy control subjects (blue line). The values represent the mean ± SEM for 15 subjects in each condition, monitored every 30 s for 7 full days.