Introduction to characteristics of normal sleep

Author: Poul J Jennum

Sleep wake regulation is undertaken by complex regulation in several parts of the brain primarily involving brain stem, thalamus, hypothalamus and basic forebrain. A primary regulatory mechanism is explained by the two-process model of sleep including homeostatic and circadian mechanism, the later primarily controlled by specific genes which are entrained by light, behaviour and food intake.

Understanding of these fundamental mechanisms are central in understanding sleep physiology, and importance for sleep disorders. In this presentation examples are presented for understanding sleep mechanism in insomnia, Parkinsonism and narcolepsy.

Data-driven topic analysis of high density EEG reveals concomitant superficial sleep during deep sleep in insomnia disorder

Authors: Julie A E Christensen1,2, Rick Wassing3, Yishul Wei3, Jennifer R. Ramautar3, Oti Kamal3, Poul J Jennum1, Eus J W Van Someren3,4,5

Affiliations: 1 Danish Center for Sleep Medicine, Department of Clinical Neurophysiology, Rigshospitalet, Glostrup, Denmark,  2 Biomedical Engineering, Department of Electrical Engineering, Technical University of Denmark, Kgs. Lyngby, Denmark, 3 Sleep & Cognition, Netherlands Institute of Neuroscience, Amsterdam, Netherlands,  4 Department of Integrative Neurophysiology, Center for Neurogenomics and Cognitive Research (CNCR), Amsterdam Neuroscience, Vrije Universtiteit Amsterdam, Amsterdam, The Netherlands, 5 Department of Psychiatry/GGZ ingeest, Vrije Universtiteit Amsterdam, Amsterdam UMC, Amsterdam, The Netherlands

Introduction: To propel the understanding of Insomnia Disorder (ID), the inconsistency between subjective and objective sleep variables needs to be resolved. We aimed to reveal sleep EEG deviations in ID using a validated analysis that quantifies co-occurrence of vigilance signatures (‘topics’).

Methods: Sleep EEG (NID=55, Ncontrols=64) was symbolized by ranking spectral power and eye-movement correlation measures within 1s bins, followed by concatenation into 3-digit sequences. A latent dirichlet allocation model uses the distribution of 3-digit sequences to describe each 30s epoch as a mixture of six topics (T1-T6), each related to a classical sleep stage. In periods with the same dominating topic for at least three epochs, we computed the probability of that topic and the co-occurrence of the remaining topics. Finally, we computed the probabilities of switching from a stable topic to any other topic.

Results: ID patients showed a higher co-occurrence of T4 (N1-related) in periods where T1 (N3-related) was dominant and stable, a higher probability of T4 in periods where T4 was dominant and stable, a higher probability of switching from a stable period of T1 to T4, and a lower probability of switching from a stable period of T1 to T2 (N3-related).

Conclusions: As compared to controls, deep sleep in ID contains more N1 sleep-related EEG signatures, suggesting continued hyperarousal during deep sleep. Where controls tend to remain in deep sleep, ID patients are more likely to switch to light sleep. Topic modeling is a powerful approach for analyzing mixtures of states.

Experiences of Cognitive Behavioral Therapy for Insomnia in Patients suffering from depression: An Interview Study

Aurthor: Henny Dyrberg, Specialpsykolog I Psykiatri, Afd. Q, Aarhus Universitetshospital, Risskov

Sleep problems are common among psychiatric patients; particularly in affective disorders. Cognitive Behavioral Therapy – Insomnia (CBT-I) is a multi-component intervention including sleep restriction, stimulus control, sleep hygiene education, cognitive therapy and relaxation techniques. Research supports the efficacy of CBT-I, but few studies focus on how patients experience the methods used in CBT-I and little is known about how depressed patients cope with the methods.

In this study 12 patients were interviewed using a semistructured interviewguide after 6 sessions of CBT-I. A qualitative content analysis was applied to analyze and interpret transcripts. Three themes were identified; “Desperate for sleep”, “I picked what, I needed” and “Staying on track”. Participants experienced hopelessness and a need to preserve energy. Behavioral techniques were experienced as hard, but effective. They chose from cognitive methods, but had problems learning too complicated methods due to cognitve problems accentuated by sleep deprivation. They used the metods in combination with behavioral methods or later as relapse prevention. Factors promoting adherence were rapid improvement, support from relatives and therapist. Factors inhibiting adherence were treatment modality, medication, timing and level of depression.

Large-scale brain networks and transition dynamics during fluctuations between wakefulness and sleep; perspectives for insomnia

 Arthor: Angus Stevns

Through sleep we are accustomed to regular and substantial modulations of consciousness. How does the brain support this modulation? Traditional models of sleep brain activity, using low-resolution scalp electroencephalography (EEG) to define a low number of homogenous sleep stages may be too simplistic to address this question. In fact, the expression of traditional EEG signatures of sleep stages is much less predictive of conscious content (or lack hereof) than previously thought. Modern neuroimaging techniques and analyses allow for the data-driven identification of dynamic large-scale network activity from continuous recordings, such as functional magnetic resonance imaging (fMRI) and magnetoencephalography (MEG). The resolution of functional connectivity networks has thus far mainly been applied to neuroimaging recording of wakefulness, with the aim of mapping large-scale signatures with relevance for on-going cognition and behaviour. An equivalent data-driven discovery of dynamic whole-brain network states in recordings of sleep may reveal spatial and temporal features of brain activity with more relevance for consciousness than traditional sleep staging.

I will present our recent advances in applying a data-driven strategy to fMRI recordings from a large group of healthy participants undergoing periods of wakefulness and NREM sleep in the scanner. The resulting decomposition of the BOLD data into whole-brain network states, defined in space and time by a Hidden Markov Model (HMM), reveals a complex transition map across the wake/NREM sleep cycle. Several aspects of the dynamic system of whole-brain networks cannot be accounted for by traditional sleep staging, but rather support a more heterogeneous view of modulations of consciousness. I will argue that such a view could be essential for a better understanding of the neural mechanisms underlying insomnia.

Sleep disturbances after severe traumatic experiences incl. Utøya massacre

Aurthor: Janne Grønli

A significant number of children and adolescents have been exposed to traumatic life events. However, knowledge about the specific sleep disturbance that occurs in individuals after trauma exposure is predominantly based on studies of adults. Grønli will present data on the long-term effects (18-30 months) on sleep in survivors after the 2011 attack at the Norwegian youth political summer camp on Utøya which resulted in the loss of 69 lives, and caused physical and mental injuries to hundreds of survivors and their relatives. In the second part of the talk, Grønli will focus on individuals who experience stress in the early life and victims of sexual abuse. Here the role of perceived social support and abuse characteristics in self-reported insomnia, nightmare frequency, and nightmare distress will be presented together with recent analyses on insomnia symptom trajectories among adult survivors of childhood sexual abuse.

European Guidelines for Diagnosis and Treatment of Insomnia

Author: Dieter Riemann Chiara Baglioni Claudio Bassetti Bjørn Bjorvatn Leja Dolenc Groselj Jason G. Ellis Colin A. Espie Diego Garcia‐Borreguero Michaela Gjerstad Marta Gonçalves Elisabeth Hertenstein Markus Jansson‐Fröjmark Poul J. Jennum Damien Leger Christoph Nissen Liborio Parrino Tiina Paunio Dirk Pevernagie Johan Verbraecken Hans‐Günter Weeß Adam Wichniak Irina Zavalko Erna S. Arnardottir Oana‐Claudia Deleanu Barbara Strazisar Marielle Zoetmulder Kai Spiegelhalder

This European guideline for the diagnosis and treatment of insomnia was developed by a task force of the European Sleep Research Society, with the aim of providing clinical recommendations for the management of adult patients with insomnia. The guideline is based on a systematic review of relevant meta‐analyses published till June 2016. The target audience for this guideline includes all clinicians involved in the management of insomnia, and the target patient population includes adults with chronic insomnia disorder. The GRADE (Grading of Recommendations Assessment, Development and Evaluation) system was used to grade the evidence and guide recommendations. The diagnostic procedure for insomnia, and its co‐morbidities, should include a clinical interview consisting of a sleep history (sleep habits, sleep environment, work schedules, circadian factors), the use of sleep questionnaires and sleep diaries, questions about somatic and mental health, a physical examination and additional measures if indicated (i.e. blood tests, electrocardiogram, electroencephalogram; strong recommendation, moderate‐ to high‐quality evidence). Polysomnography can be used to evaluate other sleep disorders if suspected (i.e. periodic limb movement disorder, sleep‐related breathing disorders), in treatment‐resistant insomnia, for professional at‐risk populations and when substantial sleep state misperception is suspected (strong recommendation, high‐quality evidence). Cognitive behavioural therapy for insomnia is recommended as the first‐line treatment for chronic insomnia in adults of any age (strong recommendation, high‐quality evidence). A pharmacological intervention can be offered if cognitive behavioural therapy for insomnia is not sufficiently effective or not available. Benzodiazepines, benzodiazepine receptor agonists and some antidepressants are effective in the short‐term treatment of insomnia (≤4 weeks; weak recommendation, moderate‐quality evidence). Antihistamines, antipsychotics, melatonin and phytotherapeutics are not recommended for insomnia treatment (strong to weak recommendations, low‐ to very‐low‐quality evidence). Light therapy and exercise need to be further evaluated to judge their usefulness in the treatment of insomnia (weak recommendation, low‐quality evidence). Complementary and alternative treatments (e.g. homeopathy, acupuncture) are not recommended for insomnia treatment (weak recommendation, very‐low‐quality evidence).

Socio-economic consequences of sleep disorders, 2018 status

Author: Poul J Jennum

Design and Validation of an End-to-End Deep Neural Network for LM and PLMS Detection

Authors: Lorenzo Carvelli1,2,*, Alexander Neergard Olesen1,2, Poul J. Jennum3, Helge B.D. Sorensen1 and Emmanuel Mignot2

Affilliations: 1 Department of Electrical Engineering, Technical University of Denmark, 2 Stanford University Center for Sleep Sciences and Medicine, 3 Danish Center for Sleep Medicine, Rigshospitalet Glostrup

A deep learning system was designed and validated for scoring PLMS and LMs during sleep. The data used in this study is the tibialis EMG polysomnography channel of 800 subjects
belonging to three datasets: Wisconsin Sleep Cohort (WSC) (275 subjects, age 54.3±7.3, BMI 32.0±6.8, %men 0.51); MrOS Sleep Study (348 subjects, age 76.2±5.8, BMI 27.3±3.8, %men
1); Stanford Sleep Clinic (SSC) (177 patients, age 44.5±14.3, BMI 27.1±6.4, %men 0.55). Manual annotations by trained technicians were used as ground truth. Each recording in WSC and SSC test subsets, totaling 60 subjects, was scored by 5 independent sleep technicians. Our algorithm was compared to each technician using the consensus of the other four as ground truth. Results show F1 score of 0.83 for WSC, 0.71 for SSC, 0.77 for MrOS. The comparison against the multiple scorers performance shows superiority of our automatic algorithm, using F1 score.

Multi-cohort validation of a new data-driven algorithm for muscular activity detection during sleep

Authors: Matteo Cesari1,2, Julie A.E. Christensen1,2, Helge B.D. Sorensen1, Friederike Sixel-Döring3,4, Claudia Trenkwalder3, Geert Mayer4, Wolfgang H. Oertel4 and and Poul Jennum2

Affiliations: 1Department of Electrical Engineering, Technical University of Denmark, Kgs. Lyngby, Denmark; 2Danish Center for Sleep Medicine, Department of Clinical Neurophysiology, Rigshospitalet Glostrup, Denmark; 3Paracelsus-Elena Klinik, Kassel, Germany; 4Department of Neurology, Philipps University, Marburg, Germany

REM sleep behavior disorder (RBD) is the main biomarker of early Parkinson’s disease (PD) and its diagnosis requires documentation of REM sleep without atonia. We developed a data-driven algorithm able to distinguish RBD patients from healthy controls (HC) and patients suffering from periodic limb movement disorder (PLMD). The algorithm identifies movements during REM and NREM sleep in chin and tibialis muscles and calculates the probabilities whether they are either in normal range, indicate RBD or PLMD. We applied the algorithm to 27 HC, 29 RBD patients, 25 PD without RBD (PD-RBD) patients, 29 PD with RBD (PD+RBD) patients from the Danish Center for Sleep Medicine and to 81 HC, 57 PD-RBD patients and 30 PD+RBD from Paracelsus-Elena Klinik, Germany. The algorithm correctly identifies normal movements in HC, and RBD movement patterns in RBD patients. PD-RBD patients show both normal and PLMD movement patterns, while PD+RBD patients show both RBD and PLMD patterns. The algorithm can be used in different cohorts to characterize movement patterns and distinguish patient groups.


Authors: Anders West, MD (1), Sofie A. Simonsen, MD (1), Henriette Sennels MD, PhD (3), Poul Jennum MD, DMSc (2), Niklas Cyril MS (1), Marie Schønsted, MB (1), Alexander Zielinski, MD (1), Birgit Sander, MSc, PhD (4), Helle K. Iversen MD, DMSc (1)
Affiliation:  (1) Clinical Stroke Research Unit; (2) Danish Center for Sleep Medicine; (3) Department of diagnostic, Clinical Biochemistry; (4) Department of Ophthalmology,
University of Copenhagen, Rigshospitalet, Glostrup

Study objectives: Natural sunlight exposure during daytime entrains the central circadian pacemaker to the 24-hour day. Given circadian control of the sleep-wake cycle, appropriately timed light exposure is crucial for optimal alertness and sleep-quality. Rehabilitation patients tend to lack exposure to sufficient natural light in daytime and are often exposed to light at unnatural time points. Installed diurnal naturalistic light may therefore positively influence the outcome of sleep quality and fatigue in this patient cohort.Methods: Stroke patients were randomized to either an intervention rehabilitation unit (IU) equipped with naturalistic lighting (artificial sunlight spectrum) or to a control rehabilitation unit (CU) with standard lighting. Time of inclusion and discharge (after at least 2 weeks) were time points for following examinations: fatigue, sleepiness, feeling of rest, subjective sleep quality, and plasma melatonin and serum cortisol levels.

Results: From 1 May 2014 to 1 June 2015, a total of 256 patients were screened, and 90 were included. At discharge, patients from the IU had less fatigue and improved rest, besides higher and evolved melatonin levels than the CU group. No differences were detected in sleepiness or sleep quality.

Conclusion: This study is the first to demonstrate that exposure to naturalistic light during admission may positively influence the circadian rhythm and fatigue in rehabilitation patients.

Funding/Disclosures: Funding: Market Development Foundation, The Capital Region grant.

Søvnapnø og sammenhæng til cerebral små-kars sygdom efter apopleksi

Author: Sofie Simonsen

Vi ved at ca 2 ud af 3 apopleksipatienter har søvnapnø og at søvnapnø resulterer i en række patologiske processer hvilket bl.a. gør at apopleksipatienter med søvnapnø har dårlige outcome og øget mortalitet. Cerebral små-kars sygdom er hyppig hos apopleksipatienter og kan beskrives ved en small vessel disease (SVD) score. Formålet med studiet var at undersøge sammenhængen mellem søvnapnø og cerebral små-kars sygdom efter apopleksi. Vi fandt i et prospektivt observationelt studie at i gruppen af patienter med behandlingskrævende søvnapnø både akut og seks måneder efter apopleksi var der signifikant højere odds ratio for at have høj end lav SVD score. Dermed konkluderer vi at cerebral små-kars sygdom ved søvnapnø bør have større opmærksomhed fremover.

Polysomnografiske indikatorer for øget mortalitetsrisiko hos apopleksipatienter

PURPOSE: The purpose of the study was to assess polysomnographic indicators of increased mortality risk in patients with stroke or a transient ischemic attack (TIA).

METHODS: We performed polysomnographies in 63 acute stroke/TIA patients. Mortality data were collected from a national database after a 19-37-month follow-up period.

RESULTS: Of the 57 stroke and 6 TIA patients, 9 stroke patients died during follow-up. All nine had moderate or severe sleep-related breathing disorders (SRBDs). Binarily divided, the group with the highest apnea hypopnea index (AHI) had an almost 10-fold higher mortality risk (hazard ratio (HR) 9.71; 95 % confidence interval (CI) 1.20-78.29; p = 0.033) compared to the patients with the lowest AHI. The patients with the longest versus shortest nocturnal wake time had a higher mortality (HR 8.78; 95 % CI 1.1-71.8; p = 0.0428). Lung disease increased mortality (HR 9.92; 95 % CI 2.00-49.23; p = 0.005), and there was a trend toward a higher mortality risk with atrial fibrillation/flutter (HR 3.63; 95 % CI 0.97-13.51; p = 0.055).

CONCLUSIONS: In stroke patients, the AHI and nocturnal wake time are indicators of increased mortality risk. SRBDs in stroke patients should receive increased attention.