Elsevier

Sleep Medicine

Volume 15, Issue 1, January 2014, Pages 125-131
Sleep Medicine

Original Article
Relationship between delta power and the electrocardiogram-derived cardiopulmonary spectrogram: possible implications for assessing the effectiveness of sleep

https://doi.org/10.1016/j.sleep.2013.10.002Get rights and content

Abstract

Objectives

The physiologic relationship between slow-wave activity (SWA) (0–4 Hz) on the electroencephalogram (EEG) and high-frequency (0.1–0.4 Hz) cardiopulmonary coupling (CPC) derived from electrocardiogram (ECG) sleep spectrograms is not known. Because high-frequency CPC appears to be a biomarker of stable sleep, we tested the hypothesis that that slow-wave EEG power would show a relatively fixed-time relationship to periods of high-frequency CPC. Furthermore, we speculated that this correlation would be independent of conventional nonrapid eye movement (NREM) sleep stages.

Methods

We analyzed selected datasets from an archived polysomnography (PSG) database, the Sleep Heart Health Study I (SHHS-I). We employed the cross-correlation technique to measure the degree of which 2 signals are correlated as a function of a time lag between them. Correlation analyses between high-frequency CPC and delta power (computed both as absolute and normalized values) from 3150 subjects with an apnea-hypopnea index (AHI) of ⩽5 events per hour of sleep were performed.

Results

The overall correlation (r) between delta power and high-frequency coupling (HFC) power was 0.40 ± 0.18 (P = .001). Normalized delta power provided improved correlation relative to absolute delta power. Correlations were somewhat reduced in the second half relative to the first half of the night (r = 0.45 ± 0.20 vs r = 0.34 ± 0.23). Correlations were only affected by age in the eighth decade. There were no sex differences and only small racial or ethnic differences were noted.

Conclusions

These results support a tight temporal relationship between slow wave power, both within and outside conventional slow wave sleep periods, and high frequency cardiopulmonary coupling, an ECG-derived biomarker of “stable” sleep. These findings raise mechanistic questions regarding the cross-system integration of neural and cardiopulmonary control during sleep.

Introduction

The biologic role of periods of nonrapid eye movement (NREM) sleep associated with low delta power is unclear. Restricting such periods has adverse consequences, including sleepiness and metabolic dysregulation similar to total sleep deprivation [1], [2], [3], [4]. Delta frequency activity on the surface electroencephalogram (EEG), usually measured between 0.5 and 4 Hz, is considered a biomarker of homeostatic sleep drive [5]. These slow waves are the scalp surface representation of a highly complex ensemble of oscillatory activity during NREM sleep, including the cortical up and down states that comprise the slow <1 Hz oscillation [6], [7], [8]. As a proportion of total EEG power, delta power is the highest during the initial cycles of NREM sleep and decreases across the biologic night, showing rebound effects following sleep deprivation [9], [10]. The frequency of occurrence of the <1 Hz slow oscillation is far greater in the first than the final NREM sleep period [11]. However, the biologic worth of stage 2 NREM sleep (N2) may not be fully accounted for by conventional scoring or by absolute delta power profiles. These disparities are especially apparent in individuals over the age of 40–50 years, in whom stage 3 NREM sleep (N3) makes up less than 20% of the sleep period [12].

We previously described a complementary approach to characterize sleep based on the interaction of autonomic and respiratory interactions (cardiopulmonary coupling [CPC]), termed the electrocardiogram (ECG)-derived sleep spectrogram. [13] Sleep spectrogram analysis reveals that NREM sleep has a distinct bimodal-type structure marked by distinct alternating and abruptly varying periods of strong high- and low-frequency CPC intensity, respectively. Much of high-frequency CPC occurs during stage N2, especially with the EEG morphology called noncyclic alternating pattern, and is associated with periods of stable breathing; a paucity of phasic EEG transients; and physiologic blood pressure dipping, which are all reduced by sleep apnea and fibromyalgia [14], [15].

Cortical slow-wave kinetics can impact autonomic and respiratory function. Slow-oscillation–like activity has been recorded in downstream neural elements (reviewed in [16]), including the hippocampus, cerebellum, thalamus, basal ganglia, and even the locus ceruleus [17]. Increased slow-wave activity (SWA) after sleep deprivation in cats was reported in subcortical structures, such as the hippocampus, amygdala, hypothalamus, nucleus centralis lateralis of the thalamus, septum, caudate nucleus, and the substantia nigra [18]. Thus cortical SWA may directly entrain activities in lower brain centers and networks, plausibly enhancing a state most conducive to the generation of sustained periods of high density slow oscillations. For example, an increased probability of a stable breathing period enabled by the slow oscillation could in turn reduce arousing respiratory afferent stimuli, thus enhancing the likelihood of undisturbed and sustained slow oscillation dense periods.

In our study, we sought to assess the relationship between EEG delta power with CPC to test the hypothesis that slow-wave power fluctuations would positively correlate with high-frequency CPC, thus identifying stage N2 periods that may have potentially similar physiologic characteristics as stage N3. Such a demonstration also would provide evidence of strong cortical effects on cardioautonomic interactions previously described for conventional slow-wave sleep (SWS) and across the entire sleep period.

Section snippets

Sleep Heart Health Study polysomnogram database

Our study used data from the Sleep Heart Health Study I (SHHS-I), a large database of home-based polysomnography (PSG) [19]. The study was designed to include ∼6000 adults at least 40 years of age, each of whom underwent at-home PSG. The study recruited participants from feeder studies in decreasing order of number of participants: Atherosclerosis Risk in Communities Study, Cardiovascular Health Study, Framingham Heart Study, New York Hypertension Cohorts, and Tucson Epidemiologic Study of

Database

Selected clinical characteristics of the subjects were as follows: mean age (±standard deviation, 60.9 ± 11 y); 63% women; body mass index (27.1 ± 4.5 kg/m2); total sleep time (TST) (366.3 ± 64.4 min); sleep efficiency (82.7 ± 10.1%); and sleep stages N1, N2, N3 + 4, and REM (%TST) (5.1 ± 3.6%, 55.6 ± 11.7%, 18.8 ± 11.8%, 20.5 ± 6.3%, respectively). The AHI was 1.8 ± 1.4 events per hour of sleep. The racial/ethnic distribution was 77% white, 8.4% black, 7.6% Native American or Alaskan, 1.8% Asian or Pacific Islander,

Discussion

The key findings of our study were (1) delta power measured from the surface EEG correlates with ECG-derived CPC high-frequency power, further supporting a link between cortical EEG electrical activity and brain stem-related cardiorespiratory functions; (2) normalized delta power provided improved correlation compared to correlations based on absolute delta power; (3) there was a consistent lag (median of approximately 4 min) between the start of the high-frequency power increase in relation to

Funding sources

Grants from the National Institutes of Health, Heart Lung and Blood Institute (RO1 HL099749 and R21 HL079248), the National Institute of General Medical Sciences and National Institute of Biomedical Imaging and Bioengineering (R01 GM104987) and the G. Harold and Leila Y. Mathers Charitable Foundation.

Conflict of interest

The ICMJE Uniform Disclosure Form for Potential Conflicts of Interest associated with this article can be viewed by clicking on the following link: http://dx.doi.org/10.1016/j.sleep.2013.10.002.

. ICMJE Form for Disclosure of Potential Conflicts of Interest form.

Acknowledgments

The Sleep Heart Health Study was supported by the National Heart, Lung and Blood Institute cooperative agreements U01HL53940 (University of Washington), U01HL53941 (Boston University), U01HL53938 (University of Arizona), U01HL53916 (University of California, Davis), U01HL53934 (University of Minnesota), U01HL53931 (New York University), U01HL53937 and U01HL64360 (Johns Hopkins University), U01HL63463 (Case Western Reserve University), and U01HL63429 (Missouri Breaks Research). A list of SHHS

References (32)

  • S. Banks et al.

    Behavioral and physiological consequences of sleep restriction

    J Clin Sleep Med

    (2007)
  • E. Tasali et al.

    Slow-wave sleep and the risk of type 2 diabetes in humans

    Proc Natl Acad Sci USA

    (2008)
  • P. Achermann

    The two-process model of sleep regulation revisited

    Aviat Space Environ Med

    (2004)
  • S. Chauvette et al.

    Origin of active states in local neocortical networks during slow sleep oscillation

    Cereb Cortex

    (2010)
  • D. Menicucci et al.

    Functional structure of spontaneous sleep slow oscillation activity in humans

    PloS one

    (2009)
  • M. Steriade et al.

    Slow sleep oscillation, rhythmic K-complexes, and their paroxysmal developments

    J Sleep Res

    (1998)
  • Cited by (65)

    • Autonomic/central coupling benefits working memory in healthy young adults

      2020, Neurobiology of Learning and Memory
      Citation Excerpt :

      Taken together, these studies suggest that experience-induced changes in SWA support efficient WM improvement. In recent years, a number of studies have explored the association between central and autonomic activity during sleep, offering promising new perspectives to understand brain-body interplay in humans (Brandenberger, Ehrhart, Piquard, & Simon, 2001; Thomas et al., 2014; Ako et al., 2003; Whitehurst, Chen, Naji, & Mednick, 2020). For example, a recent study investigated the relationship between cardiac activity and K-complexes (KC; 0.5–1 Hz)-a positive–negative-positive waveform during Stage 2 sleep similar to slow oscillations-and demonstrated that KCs were associated with a biphasic cardiac response, with a marked heart rate acceleration followed by a gradual deceleration (de Zambotti et al., 2016).

    View all citing articles on Scopus

    Deceased.

    View full text