Skip to content

Sleep and Neurodevelopment Workshop: Electrophysiologic Sleep Phenotyping (ESP)

Date/Time:

Location: Bethesda, MD

Sponsored by:
Division of Intramural Research Programs (IRP)
National Institute of Mental Health

Overview

The NIMH Division of Intramural Research Programs (IRP) convened a workshop to develop and promote Electrophysiologic Sleep Phenotyping (ESP) as a mainstay of the clinical assessment of children at risk for neurodevelopmental disorders.

Considerable gaps in knowledge exist in our understanding of sleep metrics and how these change during the early periods of neuromaturation. There is a critical need for well-designed prospective data and sample collection with integrated longitudinal follow-up to define normal vs. abnormal sleep patterns and elucidate how these patterns correlate with different risks and potential interventions.

To realize this goal, clinical scholars from pulmonary medicine, sleep medicine, neurology and child psychiatry brainstormed with psychiatric geneticists, translational scientists, epilepsy experts, child neurologists, developmental psychologists, and bioengineers. All focused on addressing the major obstacles impeding the pace of discovery in this nascent, convergent and exciting field.

Meeting Summary

Electrophysiologic Sleep Phenotyping (ESP) Project

Prior to formal talks, the group had preliminary discussions regarding the gap in data from birth to 5 years, with respect to basic understanding of neuromaturation through processes that occur during sleep.

The discussion centered around the need for systematic longitudinal data to be collected and how especially important it will be to fill the data gap on sleep neuromaturation starting in infancy, with typically developing infants and infants at risk for various neurodevelopmental disorders as very important study groups.

Perspectives on methods that can be used to maximize study retention and engagement when conducting such longitudinal studies were provided by Lonnie Zwaigenbaum, M.D, based on his experience with infant siblings of children with Autism Spectrum Disorder.

Clinical Phenomena that affect Sleep EEG interpretation

  • David Gozal, MD reviewed the literature and consensus understanding in the field surrounding the negative effects of sleep disordered breathing (SDB) on neurodevelopment in children, affecting both attention and behavioral regulation in addition to changes in the cardiovascular system and metabolic homeostasis. He also suggested that snoring alone can be used as a reliable proxy for SDB screening, and cautioned that there are most probably genetic and environmental factors contributing to which children manifest the presence of cognitive deficits at any given level of severity.
  • Greg Holmes, MD spoke to the field’s understanding of the importance of GABA for synchronization and coordination of brain oscillatory patterns and its likely outsized role in sculpting neuronal activity. In very young animals the maturation or dysmaturation of the GABA system may be reflected in the sleep electroencephalographic (EEG) coherence measures. At least one drug, bumetanide, can ‘switch’ GABA neurons from inhibitory to excitatory in animal models.

Key Points:

    • SDB should be accounted for in any standard operating procedure (SOP) at baseline and during longitudinal studies-when measuring cognitive and developmental trajectories. Respiratory parameters are often missing in extant data sets.
  • Coherence analyses by state and sleep stage associated with developmental level from the earliest possible ages are integral to the Electrophysiologic Sleep Phenotyping (ESP) project.

Panel 1: Signal Processing

  • Leila Tarokh, PhD emphasized the suitability of EEG as the central tool for this project as it is a non-invasive measure of brain activity, reflects both inter-individual variability and intra-individual stability over time and there is a strong genetic contribution to various power and density neurophysiological sleep metrics.
  • Ruth Benca, MD, PhD emphasized the dynamic nature of the sleep process, its putative integral role in brain maturation and the oft overlooked three-dimensional temporal aspects of sleep (sleep dissipation, etc.). She discussed spindle deficits, spindle topography and slow wave activity (SWA) and highlighted differences in gender. She also presented data on local EEG power reduction in subjects with obstructive sleep apnea.
  • Shafali Jeste, MD focused on the rich awake EEG research body in typical cohorts and children with ASD and emphasized a bigger potential role for EEG in biomarker discovery to identify specific genetic etiologies-- i.e., enhanced beta activity in 15 q dup (overexpression of 3 GABA receptors). She discussed divergence of chronologic age from developmental age in ASD cohorts where alpha was more closely related to development as measured by the non-verbal developmental intelligence quotient (NVDIQ) than age in months.

Key Points:

  • It will be important to establish a minimal common data elements (CDE) approach to data collection in terms of study equipment and acquisition parameters a priori, in order to best take advantage of the sleep EEG.
  • Because sleep is a local process, where you look is as important as when you look. It is essential that we establish consensus on the electroencephalogram (EEG) montage in ESP paradigms, including the minimum number of electrodes and the optimal location.
  • Given the heritability of many sleep metrics (SWS power, some spindle parameters) sleep EEG has a largely unexplored contribution to make in terms of biomarker discovery for genetic disorders.

Panel 2: Important Moderators of Sleep Physiology

  • Shaun Purcell, PhD demonstrated how the management of these very large data sets may be leveraged to glean information about relationships between clinical presentations and sleep signals. He presented aggregated spindle characteristic data on 11,630 studies across 6 cohorts to demonstrate changes in spindle density by age and frequency. Large amounts of data can move us from ‘summary data’ approaches to quantitative signal detection, including broaching broad questions about how things like the ECG signals relate to spindles, slow wave sleep (SWS), and other sleep signals, etc.
  • Katie Sharkey, MD, PhD emphasized the oft-overlooked maternal influence on circadian entrainment of the child, and on other sleep characteristics and sleep behaviors, and how those relationships likely begin in utero.
  • Mirjana Maletic-Savatic, MD, PhD pointed out that all tissues in the body respond to ‘oscillating’ metabolites in the circulation. She introduced the idea of collecting metabolites in the urine, blood and saliva at certain key points in our ESP assessment in relation to EEG maturation.

Key Points:

  • Given the enormous complexity of disentangling the genetic contribution to measurable sleep signals that may reflect neurodevelopmental/neuropsychiatric maldevelopment, we need to focus on increasing our power in pediatric sleep EEG work.
  • The maternal/fetal dyad offers valuable information pertaining to sleep mediated neurodevelopment in the very earliest periods of life.
  • Need to determine which biospecimens should ideally be part of the minimum CDE collection and at which developmental windows.

Panel 3: Translational ESP

  • Matthew McGinley, PhD introduced the idea of wakefulness phenotyping as it relates to changes of vigilance within the awake state and demonstrated that pupil diameter can be used to track changes in brain state, with fast changes tracking noradrenergics and slow changes tracking cholinergic activity. Pupillometry, pupil-indexed neuromodulation of brain physiology, can be used to sensitively track brain state, changes across sleep and wakefulness, and simple cognitive and behavioral processes.
  • Rodney Samaco, PhD reviewed genetic mouse and rat models and how these animals may be leveraged as tools to reflect various features reminiscent of human neurodevelopmental disorders (NDD). He described how the field is addressing questions of sleep abnormalities in rodent models of rare genetic syndromes as a means to understanding the broader impact of sleep abnormalities on normal CNS function. He used Rett Syndrome (RTT) as one example to illustrate: 1) how altered sleep-wake state may be linked to phenotypic outcome in NDD of known genetic etiology, and 2) how findings from multi-modal approaches may reveal new insight into the development of biomarkers and/or outcome measures for preclinical studies. Recent findings from his group support the notion that breathing abnormalities in mice deficient for the RTT-causing gene MeCP2 may be dependent on both behavioral state and age, His group is working to extend these initial findings to systematically profile and compare electrophysiologic sleep phenotypes across the lifespan of multiple genetic NDD rodent models.
  • Kiran Maski, MD, MPH talked about outcome measures potentially related to changes in REM, NREM spindles and SWS in children. Sleep neurophysiology associated with Sleep-Dependent Memory Consolidation has been the focus of her research in children. She highlighted her lab’s work with ASD and the findings that included: 1) sleep metrics correlating with memory consolidation were different for typical children (TST) than for children with ASD (slow oscillation power); 2) REM sleep parameters in ASD may inform on mood-lability and depression risks, and; 3) typical children with mild sleep apnea AHI>1 had a 50% reduction in sigma power compared to healthy kids and this reduction in sigma power predicted their memory consolidation deficits. Dr. Maski has used home sleep studies in children with ASD to measure sleep architecture. 

Key Points:

  • There is a need to validate new technologies against established gold standard sleep assessment to minimize the burden to families, cost and time delays in assessment frequency.
  • Translation models can inform greatly and need to be included more centrally in discussions about the intersection of sleep & neurodevelopment.
  • The relative causal contributions, moderating and mediating effects of the REM/NREM/SWA/sleep spindles to learning and memory in children have not been thoroughly worked out in the typically developing brain. Yet, we have suggestions from the limited extant pediatric work that these metrics may be used to help separate diagnostic groups and show relationships between sleep and cognitive and behavioral outcomes.

The National Sleep Research Resource (NSRR)

Susan Redline, MD, MPH highlighted that the NSRR data repository and community engagement platform is designed to bring together many sets of data while providing the user ability to visualize, query, and analyze information. The need for aggregating data is to overcome current limitations of datasets with insufficient sample size, diversity, and/or breadth of clinical and biological variables. Features of NSRR were highlighted that are designed to promote collaboration and better use of physiological signal data, such as including mechanisms for sharing signal processing tools, providing clear annotations of signals from diverse sources, and controlled data vocabularies (supported by canonical data dictionaries) to aid the user in understanding the definitions and provenance of key variables. NSSR has the potential to speed the pace of knowledge discovery and foster research collaborative opportunities for secondary analyses, genotype-phenotype analyses with links to other resources (e.g. dbGap), and signal analysis tool development and validation. In addition to NSRR, Dr. Redline’s team has leveraged related informatics tools to support prospective multi-center data collection of physiological signals and a wide variety of neurobehavioral and other information.

In addition, the workshop discussants determined that there is utility in identifying and evaluating existing datasets for use in validating signals, generating hypotheses, and generating preliminary data to inform prospective initiatives. 

Key points:

  • Large, prospective, longitudinal efforts like ESP may be accelerated by leveraging the NSRR platform. This platform can be used to support both prospective data collection as well as aggregate and organize data from existing clinical and research data sources.
  • NSRR has not yet been leveraged for neurodevelopmental/neuropsychiatric populations specifically, creating an opportunity for Sleep & Neurodevelopment Consortium and ESP.
  • NSRR and other data sources will be instrumental in allowing for comprehensive analysis of retrospective datasets.

Conclusions

The multidisciplinary group concluded with the imperative to establish a standardized protocol to acquire and analyze pediatric sleep EEG signals in association to neurodevelopmental milestones. Normative data are essential for understanding the basis of sleep and neurodevelopment. A consortium is established to create and initiate standardized protocols for collecting longitudinal, prospectively ascertained electrophysiologic sleep data with contemporaneous behavioral phenotyping to identify normal and abnormal developmental trajectories amenable for potential therapeutic intervention. The potential here lies in the ability to track sleep-dependent neuromaturation in the developing brain in a longitudinal, prospective manner, thereby establishing both normal and aberrant trajectories.

Affiliations of Workshop Speakers and Discussants:

Ruth Benca, MD, PhD
Department of Psychiatry and Human Behavior
University of California, Irvine 

Ashura Buckley, MD
Sleep and Neurodevelopment Service
Office of the Clinical Director
National Institute of Mental Health

Daniel G. Glaze, MD
The Children’s Sleep Center
Texas Children’s Hospital and Department of Pediatrics
Baylor College of Medicine

Robin Goin-Kochel, PhD
Autism Center
Texas Children's Hospital 

David Gozal, MD, MBA
Department of Pediatrics
The University of Chicago

Deborah Hirtz, MD
Department of Neurological Sciences
University of Vermont College of Medicine 

Gregory L. Holmes, MD
Department of Neurological Sciences
University of Vermont College of Medicine 

Shafali Jeste, MD
Jeste Lab - Center for Autism Research and Treatment at UCLA
University of California, Los Angeles

Mirjana Maletic-Savatic, MD, PhD
Jan and Dan Duncan Neurological Research Institute
Texas Children's Hospital 

Omar Khan, MD
National Institute of Neurological Disorders and Stroke

Monique LeBourgeois, PhD
University of Colorado, Boulder 

Kiran Maski, MD, MPH
Boston Children's Hospital
Department of Neurology 

Matthew McGinley, PhD
Duncan Neurological Research Institute and Baylor College of Medicine 

Snezana Milanovic, MD, MSc
Mother-Child Wellness Clinical and Research Center
Boston Medical Center
Boston University 

Maryland Pao, MD
Office of the Clinical Director
National Institute of Mental Health 

Shaun Purcell, PhD
Department of Psychiatry
Brigham and Women’s Hospital 

Susan Redline, MD, MPH
Brigham and Women's Hospital
Harvard Medical School 

Alcibiades Rodríguez, MD
NYU Langone Medical Center
New York University 

Jared Saletin, PhD
Psychiatry and Human Behavior
Brown University 

Katie Sharkey, MD, PhD
Departments of Medicine and Psychiatry & Human Behavior
Brown University

Lin Sikich, MD, MA
Duke Center for Autism and Brain Development 

Rodney Samaco, PhD
Jan and Dan Duncan Neurological Research Institute
Texas Children's Hospital 

Sophie Schwartz
Boston University Center for Autism Research Excellence 

Leila Tarokh, PhD
University Hospital of Child & Adolescent Psychiatry and Psychotherapy Research Department
University of Bern, Switzerland

Audrey Thurm, PhD
Neurodevelopmental and Behavioral Phenotyping Service
Office of the Clinical Director
National Institute of Mental Health 

Michael Twery, PhD
Director, National Center on Sleep Disorders Research
National Heart, Lung, and Blood Institute

Phyllis Zee, MD, PhD
NHLBI Council and Department of Neurology
Northwestern Memorial Hospital

Lonnie Zwaigenbaum, MD
Department of Psychiatry
University of Alberta, Canada