How stressful are 105 days of isolation? Sleep EEG patterns and tonic cortisol in healthy volunteers simulating manned flight to Mars

https://doi.org/10.1016/j.ijpsycho.2014.04.008Get rights and content

Highlights

  • Six subjects were studied during a 105 day spaceflight simulation.

  • Diurnal stress levels and night sleep EEG were monitored during the simulation.

  • Higher cortisol yielded a shortened and fragmented sleep with a REM latency decrease.

  • These changes were associated with a delta activity decrease and a beta increase.

  • Delta activity at higher cortisol levels had a left hemispheric prevalence while beta lateralized on the right.

Abstract

Spaceflights “environment” negatively affects sleep and its functions. Among the different causes promoting sleep alterations, such as circadian rhythms disruption and microgravity, stress is of great interest also for earth-based sleep medicine. This study aims to evaluate the relationships between stress related to social/environmental confinement and sleep in six healthy volunteers involved in the simulation of human flight to Mars (MARS500).

Volunteers were sealed in a spaceship simulator for 105 days and studied at 5 specific time-points of the simulation period. Sleep EEG, urinary cortisol (24 h preceding sleep EEG recording) and subjectively perceived stress levels were collected. Cognitive abilities and emotional state were evaluated before and after the simulation. Sleep EEG parameters in the time (latency, duration) and frequency (power and hemispheric lateralization) domains were evaluated.

Neither cognitive and emotional functions alterations nor abnormal stress levels were found. Higher cortisol levels were associated to: i) decrease of sleep duration, increase of arousals, and shortening of REM latency; ii) reduction of delta power and enhancement of sigma and beta in NREM N3; and iii) left lateralization of delta activity (NREM and REM) and right lateralization of beta activity (NREM).

Stressful conditions, even with cortisol fluctuations in the normal range, alter sleep structure and sleep EEG spectral content, mirroring pathological conditions such as primary insomnia or insomnia associated to depression. Correlations between cortisol fluctuations and sleep changes suggest a covert risk for developing allostatic load, and thus the need to develop ad-hoc countermeasures for preventing sleep alterations in long lasting manned space missions.

Introduction

Poor sleep quality has been frequently described as a common response to stress or distress (Van Reeth et al., 2000). The exposure to chronic stress often determines alterations of sleep, affecting both NREM and REM phases. NREM changes consistently observed in human studies are the increase of shallow sleep and the reduction of Slow Wave Sleep (SWS or NREM stage N3). REM changes are the increase of its duration and the shortening of the latency of the first REM period, and a general increase of REM density (Van Reeth et al., 2000). Studies both on animal models of chronic stress (Cheeta et al., 1997, Opp, 1995) and on humans (Opp, 1995) suggest that these alterations are likely mediated via corticotropin-releasing hormone (CRH) release, which in turn stimulates endogenous adreno-cortico-tropic hormone (ACTH) and cortisol secretion (i.e. hypothalamic pituitary adrenal axis – HPA – activation). HPA hyperactivation related to chronic stress is paralleled by a right hemisphere hyperarousal during wake (Hewig et al., 2008).

Astronauts involved in real or simulated long-term missions represent a model for the study of chronic stressful engagement with “real or simulated extreme environment” (Monk et al., 1998). Reduction of SWS and REM sleep, shortening of REM latency and increase of arousals are typically observed in short-term space missions, due to circadian rhythm phase shifts, heightened workload demand, emergencies, and more in general, extraordinary stressful environmental conditions (Dijk et al., 2001, Mallis and DeRoshia, 2005). In this light, a crucial problem for long lasting manned missions in space is represented by sleep alterations. Sleep EEG data from long-term space missions are scarce and ground-based simulation of long lasting missions is at present the experimental condition of choice for the study of such alterations. Recently, a study investigating the relationships between sleep and long lasting spaceflight simulation has been published (Basner et al., 2013). This study, based on actigraphy and neurobehavioral assessment, identified that the simulation of long lasting space mission induces alterations of sleep quality and of sleep-wake periodicity and timing, which are mainly sustained by an inadequate circadian entrainment.

Here we report a study in which the model of stress is represented by the simulation of the human flight to Mars (MARS105, pilot phase of the MARS500 project). Six healthy volunteers were kept for 105 days within an isolation facility mimicking a spacecraft and were studied during the flight simulation as well as right before and after the period of confinement. In this condition, stress was represented by social and environmental confinement, enforced interaction between crewmembers, resources rationing, limited and lagged communications with the outside and environmental acoustic noise.

The working hypothesis of the study addresses the issue of stress vulnerability to spatial confinement and social isolation in a super-selected and well-trained population. This pilot study is the first international attempt aimed at identifying indices of abnormal stress responses in healthy volunteers in order to design and develop ad hoc countermeasures to be administered in the study of 520 days of isolation (MARS 500), which represents the real simulation of human flight to Mars. The aim of the study was to examine how stress, measured by tonic cortisol levels and subjectively perceived stress levels, affects sleep in terms of structure and EEG power/lateralization and whether it produces long lasting effects on cognitive abilities and emotional state that persist beyond the confinement duration.

As in many simulation studies with such complex experimental setups (Basner et al., 2013, Schneider et al., 2010), the choice of an appropriate statistical analysis has been conditioned by two important issues concerning the high number of parameters necessary for sleep characterization and the involvement of a small number of participants and sampling times. In order to override these intrinsic limitations, a hierarchical approach aimed at pruning the numerous variables that characterize sleep patterns was employed in the data analysis. Principal Component Analysis (PCA) and confirmatory ANCOVAs performed on the selected variables allowed the identification of significant associations between specific sleep features and cortisol levels, considered by consensus a reliable indicator of stress.

Section snippets

Subjects selection and experimental setup

Six healthy right-handed male volunteers (mean age 33 ± 6 years) have been selected for a 105-days (from March 31st to July 14th, 2009) mission to Mars simulation organized by the Institute for Biomedical Problems (IBMP, Moscow, Russia) and the European Space Agency (ESA). Crewmembers selection procedures, based on a thorough psychophysical evaluation of the candidates, were the same of those used for real astronauts (for further details on the selection procedure see Supplementary material). The

Cognitive and emotional neuro-psychological evaluation

The comparison of psychometric scores did not reveal significant differences between BDC and RDC phases either for cognitive abilities or for emotional state (see Table 1).

Stress and sleep time-domain parameters

Table 2 summarizes descriptive statistics of stress and sleep time-domain parameters related to each subject. Note that crewmembers showed on average: (i) perceived stress scale (PSS) scores lower than normative levels (Cohen and Williamson, 1988); (ii) cortisol levels within the normality range (Görges et al., 1999), and

Discussion

We investigated the relationships between stress, sleep, cognitive abilities and emotional state in six male volunteers enrolled in the MARS500 pilot phase and sealed within an isolation facility module for 105 days.

The principal result of this study is the identification of relationships between stress markers and specific sleep macrostructure features as well as EEG rhythms.

Crewmembers showed perceived stress scale (PSS) scores lower than normative levels (Cohen and Williamson, 1988) and

Acknowledgments

This work is dedicated to the memory of Prof. Mario Guazzelli.

This research was supported by Italian Space Agency (ASI) through grant ASI no. I/006/09/0

We thank the Institute of Biomedical Problems (IBMP, Moscow, Russia) and the European Space Agency (ESA) for the organization of the 105 day simulation.

We thank Luca Bastiani and Marco Scalese of Social Epidemiological Section of the Institute of Clinical Physiology of the National Research Council for revising the statistical procedures.

References (58)

  • M.R. Opp

    Corticotropin-releasing hormone involvement in stressor-induced alterations in sleep and in the regulation of waking

    Adv. Neuroimmunol.

    (1995)
  • L. Palagini et al.

    REM sleep dysregulation in depression: state of the art

    Sleep Med. Rev.

    (2013)
  • G.R. Poe et al.

    Cognitive neuroscience of sleep

    Prog. Brain Res.

    (2010)
  • S. Schneider et al.

    Exercise as a countermeasure to psycho-physiological deconditioning during long-term confinement

    Behav. Brain Res.

    (2010)
  • A. Steiger

    Neurochemical regulation of sleep

    J. Psychiatr. Res.

    (2007)
  • A. Steiger et al.

    Wake and sleep EEG provide biomarkers in depression

    J. Psychiatr. Res.

    (2010)
  • O. Van Reeth et al.

    Interactions between stress and sleep: from basic research to clinical situations

    Sleep Med. Rev.

    (2000)
  • M. Vandekerckhove et al.

    The emotional brain and sleep: an intimate relationship

    Sleep Med. Rev.

    (2010)
  • D. Andrew et al.

    NREM sleep EEG frequency spectral correlates of sleep complaints in primary insomnia subtypes

    Sleep

    (2002)
  • I.A. Antonijevic et al.

    Hyporesponsiveness of the pituitary to CRH during slow wave sleep is not mimicked by systemic GHRH

    Neuroendocrinology

    (1999)
  • P.T. Barrett et al.

    The observation to variable ratio in factor analysis

    Pers. Study Group Behav.

    (1981)
  • M. Basner et al.

    Mars 520-d mission simulation reveals protracted crew hypokinesis and alterations of sleep duration and timing

    Proc. Natl. Acad. Sci. U. S. A.

    (2013)
  • K.A. Bennion et al.

    Sleep and cortisol interact to support memory consolidation

    (2013)
  • A.A. Borbély et al.

    Effect of benzodiazepine hypnotics on all-night sleep EEG spectra

    Hum. Neurobiol.

    (1985)
  • R.B. Cattell

    The Scientific Use of Factor Analysis

    (1978)
  • S. Cohen et al.

    A global measure of perceived stress

    J. Health Soc. Behav.

    (1983)
  • S. Cohen et al.

    Perceived stress in a probability sample of the United States

  • T.T. Dang-Vu et al.

    Interplay between spontaneous and induced brain activity during human non-rapid eye movement sleep

    Proc. Natl. Acad. Sci. U. S. A.

    (2011)
  • G. Dèchamps et al.

    Longitudinal assessment of psychological adaptation during a winter-over in Antarctica

    Environ. Behav.

    (2005)
  • Cited by (0)

    View full text