Sleep Stage Moderation Using Automated Auditory Stimulation
Overview
The study introduces a pioneering protocol for the automated detection and suppression of slow-wave sleep (SWS), a crucial physiological process with implications for both basic science and clinical applications. The protocol involves the automated identification of SWS through the entire head topography of frontal slow waves and its subsequent suppression using closed-loop modulated randomized pulsed noise. The research, conducted on 15 healthy young adults in a repeated-measure sleep laboratory study, compares the efficacy, feasibility, and functional implications of the automated protocol to sham stimulation.
Key findings
1. Auditory Stimulation Significantly Reduces SWS
– Auditory stimulation, compared to sham stimulation, resulted in a highly significant 30% reduction in slow-wave sleep without impacting total sleep time.
2. Effects on Sleep Architecture
– The diminished SWS was accompanied by an increase in lighter non-rapid eye movement sleep.
– Slow-wave activity shifted towards the end of the night, suggesting a homeostatic response and functional relevance.
3. Cumulative Slow-Wave Activity Reduction
– Cumulative slow-wave activity across the night was significantly reduced by 23%, indicating the effectiveness of the automated suppression protocol.
4. Impact on Wake Electroencephalographic Theta Activity
– Undisturbed sleep naturally led to a reduction in wake electroencephalographic theta activity from evening to morning, associated with synaptic downscaling during SWS.
– Suppression of SWS prevented this dissipation, revealing a notable impact on neurophysiological processes.
5. Feasibility and Efficacy of the Automated Protocol
– The study provides evidence supporting the feasibility, efficacy, and functional relevance of the novel fully automated SWS suppression protocol based on auditory closed-loop stimulation.
6. Call for Future Research
– The researchers emphasize the need for further investigations to explore the functional relevance and potential clinical applications of this automated SWS suppression protocol.
In summary, the study introduces an innovative automated protocol for SWS suppression, demonstrating its effectiveness, feasibility, and potential impact on sleep architecture and neurophysiological processes. Further research is encouraged to delve into the functional implications and broader applications of this novel approach.
Introduction
The study addresses the need for innovative protocols in the automated modulation of slow-wave sleep (SWS), a vital aspect of non-rapid eye movement (NREM) sleep with significant implications for both basic science and clinical applications. While recent advancements have focused on enhancing SWS, the development of protocols for automated SWS suppression remains understudied. SWS manipulation has potential clinical applications, including antiepileptic and antidepressant effects. Notably, therapeutic sleep deprivation, involving SWS suppression, has shown promise in rapidly improving depressive symptoms in patients with major depressive disorder.
The physiological basis of SWS involves slow, high-amplitude rhythmic fluctuations (slow waves or SWs) in electroencephalographic (EEG) recordings. Current online SWS state detection relies on the spectral signature of SWs, specifically slow-wave activity (SWA), which reflects the homeostatic regulation of sleep–wake cycles. On a neural level, SWs indicate a bistability in the resting membrane potential of cortical neurons. EEG topographically, SWS is characterized by distinctive maps reflecting frontal dominance corresponding to different brain states.
Fully-automated SWS suppression faces challenges due to inter-individual differences and the homeostatic nature of SWS. Reliable SWS detection requires addressing these challenges. The study introduces a novel protocol for the fully-automated online detection and suppression of SWS in healthy individuals. The approach involves real-time detection based on a topographic template of frontal SWs and a stimulation protocol using closed-loop modulated random bursts of pink noise. Acoustic stimulation has proven effective and safe for SWS modulation.
The primary objective of the study is to demonstrate SWS suppression through the novel protocol compared to a sham night in a repeated-measures sleep laboratory study. The evaluation includes assessing the time course of SWA across the night and wake EEG theta activity as indices of functional relevance. SWA is considered a marker and modulator of overall synaptic strength across the sleep–wake cycle, aligning with the synaptic homeostasis hypothesis. EEG theta activity serves as a correlate of SWA during wakefulness.
The study contributes to the field by introducing an automated approach to SWS suppression, addressing limitations associated with manual or semi-automated protocols. The findings provide initial evidence supporting the feasibility and efficacy of the proposed protocol, opening avenues for further research into the functional implications and clinical applications of automated SWS modulation.
Methods
The study involved 18 healthy university students, with 15 participants included in the final analysis (eight females, mean age 23.5 years). The SWS detection algorithm was developed based on data from a second cohort of 12 healthy university students. Participants provided written informed consent, and the study adhered to ethical standards.
The study employed a repeated-measures design with three sleep laboratory nights: one adaptation night and two experimental nights with auditory stimulation and sham (counterbalanced order). EEG was recorded using a 128-channel EEG cap, and participants completed questionnaires assessing various factors. A psychomotor vigilance task (PVT) evaluated morning vigilance, and wake resting-state EEG theta activity was recorded.
EEG data were pre-processed, and slow-wave sleep (SWS) was detected using a topographic template of frontal slow waves. Auditory stimulation was applied through earphones, and the study employed statistical analyses, including Wilcoxon signed-rank tests for non-normally distributed sleep architecture data.
The primary hypothesis was that stimulation would reduce time spent in SWS. Various sleep architecture variables were assessed, including total sleep time, sleep efficiency, sleep onset latency, wake after sleep onset, time in different sleep stages, and the arousal index. The study also analyzed EEG spectral power densities, slow-wave energy (SWE), and the impact of stimulation on evening-to-morning reduction of wake theta power.
Results indicated a reduction in SWS time with auditory stimulation. Sleep architecture variables and EEG spectral analyses were used to evaluate the impact of stimulation on different sleep stages and neural activity. The study provided a comprehensive analysis of the effects of auditory stimulation on sleep parameters, vigilance, and EEG activity, contributing valuable insights into automated SWS modulation protocols.
Result
The study revealed a significant reduction in slow-wave sleep (SWS) by 29.6% with auditory stimulation compared to sham (p = 0.0024; effect size d = 0.90). This reduction was accompanied by an increase in sleep stage N2 (11.8%, p < 0.001; d = 1.02) and a decrease in REM sleep (12.3%; p = 0.027; d = 0.66). Vigilance state-transitions demonstrated a significant decrease in the continuity of SWS (15.0%; p < 0.0001) and reduced continuity of N2 sleep (4.6%; p = 0.0058). Additionally, there was a decreased transition probability from wake to REM sleep (67.0%; p = 0.018).
EEG spectral power analyses aligned with expectations, showing a significant reduction in slow-wave activity (SWA) across the scalp during non-rapid eye movement (NREM) sleep with stimulation (23.1%; p < 0.01; d = 0.98). This reduction was primarily driven by the suppression of SWA during SWS (21.6%; p < 0.01; d = 0.77). Conversely, SWA increased in lighter NREM sleep stages with stimulation (17.0%; p < 0.01; d = 1.01). End-of-night slow-wave energy (SWE) across the scalp was significantly reduced with stimulation (22.6%; p < 0.01; d = 0.92), indicating that the increased SWA during lighter NREM sleep did not offset the suppression during SWS.
Analyses of slow waves (SWs) revealed a significant reduction in SW count (18.1%; p = 0.045; d = 0.57), amplitude (11.7%; p = 0.012; d = 0.74), and density (18.6%; p = 0.029; d = 0.63) during NREM sleep with stimulation. This reduction was driven by selective suppression during SWS for SW count (45.0%; p < 0.001; d = 1.14) and amplitude (5.3%; p = 0.037; d = 0.59). SW density was reduced on a trend level (18.9%; p = 0.056; d = 0.54). Conversely, SW amplitude increased during lighter NREM sleep (12.3%; p = 0.011; d = 0.76).
While the stimulation protocol did not significantly affect spindle events during NREM sleep, it led to a reduced number (25.8%; p = 0.012) and density (26.1%; p = 0.013) in SWS. In contrast, both the number (14.5%; p < 0.01) and density (13.4%; p < 0.01) of spindle events increased in lighter NREM sleep.
Wake EEG theta power analysis indicated an interaction effect between stimulation condition and time of day (p = 0.017, F = 7.4). Post hoc comparisons revealed a decrease in theta power from evening to morning in the sham night only (14.7%; p < 0.01). Theta power in the morning was significantly higher after stimulation compared to the sham night (13.3%; p < 0.01).
Behavioral data, including psychomotor vigilance task (PVT) measures and subjective reports, showed no significant differences between stimulation and sham nights, indicating that the stimulation was well-tolerated. Participants reported hearing tones more often during the stimulation night but had similar hearing threshold values. The study provides comprehensive insights into the impact of auditory stimulation on sleep parameters, EEG activity, and vigilance states.
Despite no significant impact on spindle events during NREM sleep, stimulation led to a reduced number (25.8%; p = 0.012) and density (26.1%; p = 0.013) in SWS. Conversely, both the number (14.5%; p < 0.01) and density (13.4%; p < 0.01) of spindle events increased in lighter NREM sleep.
Analysis of wake EEG theta power revealed an interaction effect between stimulation condition and time of day (p = 0.017, F = 7.4). Post hoc comparisons disclosed a decrease in theta power from evening to morning in the sham night only (14.7%; p < 0.01). Theta power in the morning was significantly higher after stimulation compared to the sham night (13.3%; p < 0.01).
Behavioral data, encompassing psychomotor vigilance task (PVT) measures and subjective reports, indicated no significant differences between stimulation and sham nights, affirming the well-tolerated nature of the stimulation. Participants reported hearing tones more frequently during the stimulation night but exhibited similar hearing threshold values. This study furnishes comprehensive insights into the effects of auditory stimulation on sleep parameters, EEG activity, and vigilance states.
Conclusion
In this study, a fully automated auditory closed-loop protocol was designed and assessed for selectively suppressing slow-wave sleep (SWS) in healthy adults. The protocol utilized a novel SWS detection method based on a topographic template of frontal slow waves and a randomized pulsed-noise-stimulation protocol with closed-loop volume modulation to suppress SWS. The protocol exhibited robust performance, delivering 95% of stimuli during SWS or N2 sleep, leading to a significant 30% reduction in SWS across non-rapid eye movement (NREM) sleep compared to the sham night, with a large effect size.
The SWS detection approach, rooted in recent work on topography-based slow wave-phase detection, demonstrated reliability in identifying SWS, particularly when combined with a parallel fast arousal/artifact detection process. This allowed for a selective suppression of SWS across participants and the entire night without compromising total sleep time.
The stimulation protocol, employing closed-loop modulated random bursts of pink noise, aimed to overcome challenges observed in previous studies where increasing volume led to arousal pathways activation. The randomized features of the stimuli, including duration and inter-stimulus intervals, introduced unpredictability in volume, potentially enhancing effectiveness. The study’s primary finding was a significant suppression of SWS in the stimulation night, accompanied by a shift into lighter NREM sleep (N2), aligning with prior research on SWS suppression. Importantly, there was no significant increase in arousal or wakefulness.
Analysis of slow-wave activity (SWA) revealed a disruption in its typical time course, indicating a functional relevance and homeostatic response. The inhibition of the dissipation of wake EEG theta power, correlated with the reduction in SWS and total numbers of slow waves, suggested potential implications for synaptic downscaling during sleep. Behavioral measures, including psychomotor vigilance task (PVT) and subjective reports, showed no significant alterations, emphasizing the short-term compensatory capacity of healthy participants.
The study’s implications extend to both basic science and clinical applications. The automated protocol offers a tool for investigating the functions of SWS, such as memory formation, synaptic reorganization, and metabolic clearance. Moreover, selective SWS suppression, as facilitated by this protocol, may have clinical relevance, especially in conditions like major depressive disorder (MDD). The approach opens avenues for testing and refining therapeutic models, potentially providing a less demanding and more sustainable alternative to traditional sleep deprivation methods in clinical settings. The automated nature of the SWS detection approach enhances its applicability across diverse age groups and patient populations, eliminating the need for constant experimenter supervision. Overall, the study demonstrates the feasibility, efficacy, and functional relevance of the developed SWS detection and suppression protocol, paving the way for potential broader applications and future treatment developments.
Oncology Related Tools
Other
Latest Research
